Innovation in Transportation: Proceedings of a Workshop, September 24-26, 1979, National Academy of Sciences, Washington, D.C. (1980)

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.

REMARKS BY CHAIRMEN ON THE SCOPE OF PANEL DELIBERATIONS, PRESENTATION OF BACKGROUND PAPERS, AND DISCUSSANTS' COMMENTS 45

THE SETTING FOR INNOVATION REMARKS BY FOSTER L. WELDON My remarks will be rather brief. I plan only to outline a framework within which I think our panel can approach its assignment. In the process, I hope to suggest how we might view the setting for innovation as a starting point for our deliberations. People look at the term "setting" in different ways. For my own part, I would like to define setting as simply the environment within which a transportation change might take place. What I am hoping we can do is look at transportation needs in terms of performance require- ments and explore, then, why our abundant technology has not been appli- ed more effectively, instead of looking at what hardware might have been applied. In other words, we want to get away from specific hardware ideas in order to explore the larger setting: why technology has not been applied more fruitfully. I know I will get some arguments on this point, because many people say one cannot really look at the environment for innovation ex- cept in a specific problem context. That makes good sense. Certainly an innovative solution to an air-scheduling problem is developed in an environment that is quite different from that in which a productivity improvement in a marine terminal is made. But I have a rather simple-minded answer to that. If we imagine absolutely the worst environment for transportation innovation and if we suggest ways for improving that setting, then I think we will have some results that are generally applicable. That is what we will be trying to do, and, of course, the model "worst" environment that was in the back of my mind when I developed this discussion framework is the urban transportation setting. That is where the diffuse trip problems are. That is where most of the politi- cal problems are. Therefore, how does one categorize a worst environment from the point of view of looking at the setting for transportation innovation? I have selected four major headings that I believe cover everything we need to discuss. Number one is the governmental setting. I isolated that one because certainly almost everything that is planned or done in transportation is affected by or impinges on government in one way or another and at one or more levels. 47

Number two is the industrial-commercial setting. Industrial- commercial entities develop and provide almost all transportation equip- ment and services, so the environment there needs a clear look. Number three is the research setting. Why put that in? Well, research certainly is the foundation for long-range planning and develop- ment in transportation, and we need to take a good look at that side of the problem. The fourth heading is the implementation setting, not because it is distinct from the three already listed but because it will permit us to focus on some very refractory problems that are common to the other three categories. So, we have all these components to look at: governmental, indus- trial-commercial, research, and the catchall, implementation settings. Our overall objective, in looking at innovation in this framework, is to see if we can identify some recommended changes that will help DOT expedite the innovative process. The governmental setting necessarily includes the federal govern- ment, state governments, and local governments—all those city, county, township, and special purpose districts or authorities that are set up to monitor or to operate transportation. The federal level appears to be the best source of funding for innovative programs, but unfortunately, the federal government is not the real customer for transportation innovation. Unfortunately again, the local arena, which really is the customer in almost every case, is a hodgepodge of all sorts of different quasi-governmental agencies that all have different ideas about what is good for them. At last count, in our 200-plus standard metropolitan statistical areas, there were more than l8,000 governmental units—a highly fractionated customer indeed to be convinced and compromised into accepting any innovative idea in trans- portation. What types of questions will we be asking about this governmental setting? A few examples: Is there any way to stimulate innovation at the local level simply through a judicious choice of initial projects? Demonstration programs have tried time and time again, but too often by the time a project gets approved it has been so compromised to accom- modate conflicting viewpoints that it does not represent innovation at all, and by the time the project is in place it contains nothing more than off-the-shelf components and concepts. Other obvious questions: How can federal resources best be deploy- ed to promote innovation? Not just through funding alone but perhaps through basic research? How about the state's role? What is it now? More or less a pass-through agency for funds? What should it be? We have a lot to look at under this heading. In the industrial-commercial setting, there are at least four factors we must consider—the equipment suppliers; the transport system operators; the architectural, engineering, and construction firms; and the special interest organizations, that is the professional, occupa- tional, and industrial associations that represent the others. All are important in the innovation scene, and we want to find out what their influences are. 48

One thing that needs to be pointed out is that the equipment suppliers and the transport system operators certainly represent a rather mature industry that is heavily invested in fixed plant. In other words, they have characteristics that are not particularly conduc- ive to entrepreneurial or agile innovation. We want to examine whether this is indeed a deterrent to innovation and what might be done to change the situation. The architectural, engineering, and construction firms, (A-E and C companies), though not similarly burdened with fixed plant, may also have a built-in inertia to change that derives from all of the rigid standards under which they must operate. Just building codes, for example, and con- struction standards may create significant deterrents to innovation. The special interest organizations and trade associations are generally dedicated to status quo, I believe, to protect the interests of their membership. So there is a lot of inertia in all these areas, and the kind of question we will be asking is, what can DOT do to help overcome this resistance to change? As to the research setting, at least four types of research facili- ties enter into the picture. These are the government R§D facilities, the industrial R§D facilities, the academic research centers, and the independent research organizations. Government R§D certainly houses a great deal of research talent and facilities, but I do not believe that transportation is really getting a fair share of the spinoff from all of these resources. Very little, in my experience, has been brought directly to bear on transportation problems from this source. Theoretically, the industrial R§D actitivities are available to DOT through the request for proposal (RFP) process, but here, again, there are some serious problems. Many companies are reluctant to par- ticipate in bidding for a number of reasons, e.g., contract restrictions on the direction of effort, allowable costs. In the academic centers, I detect a considerable decline in inno- vative transportation activity. There is really no transportation curriculum in most institutions, and transportation centers themselves, in some cases, at least, are suffering from what all universities are going through now, declining enrollment and increasing costs. The first cuts are bound to come in the nondepartmental activities of the university. So that leaves the independent research organizations that are specially well-organized to handle the RFPs and respond to government proposals. A lot of good work is done in this sector, but that sort of activity does not fulfill the university role of producing young, inno- vative talent to go into industrial transportation activities. The fourth category is the setting for implementation; without implementation there is, of course, no innovation. Certain factors here are particularly important. One is the physical system character- istic of transportation. This inhibits any kind of innovation, or so it is said, just by virtue of its size. It is massive; it is complex^ and how can one change it significantly in any reasonable time frame? This is one aspect of the implementation setting that I would like to look at very critically. 49

Labor-management attitudes and objectives also need to be singled out and looked at very carefully. Associated with these are the mar- ket characteristics and certain human and organization factors that inhibit innovation. One final point in regard to the implementation of innovative trans- portation concepts is the terrible dilemma that faces a transportation innovator in the private sector. He cannot really risk massive company funds to test a system for which he has no measure of the down-the-road payoff. Certainly, it would be possible to construct fancy demand models to predict performance for expanding present transportation sys- tems, but for a really innovative system change there are no hard data to plug into the model short of building and testing the new concept. A prudent manager simply cannot put money into that kind of thing. One cannot afford to put a good idea into action just to get the data one needs to evaluate the risks of the idea as an ongoing commercial venture. This business of the speculative nature of transportation innova- tion leads directly into labor-management questions as well. Risking front-end money is only a small part of the picture in proceeding with a transportation innovation. There is the risk of upsetting the esta- blished labor-management relationships that are the foundation of the transportation business that one had before the innovation. So, the prospective change could mean risking more than the initial cost of in- novation; it could mean risking one's whole business, because a strike could shut down and even ruin it. Aside from labor problems, there are organizational and human factors. Organizations and people are uncomfortable with change. If things are going well, why rock the boat? And so it is the very com- panies that can afford innovation that are least likely to try it because they are doing all right anyway and they could put the front- end money back into their existing business at no risk and probably make out just as well. INNOVATION AND THE STRUCTURE OF TRANSPORTATION ACTIVITIES BY WILLIAM L. GARRISON Our thesis is structural and deterministic — innovation in transportation is constrained by the structures of transportation activities that pro- vide the environments for innovation and its adoption. Innovation and technology supply, in turn, affect industry structure. In addition, transportation activities adhere to development paths that may be describ- ed as growth "dynamics," patterns, or cycles; innovation opportunities and impacts differ upon the growth dynamic circumstances. We begin our analysis of transportation innovation by describing the principal features of transportation activities and characterizing innovation and technology deployment activities. The discussion then narrows to the analysis of the characteristics of the separate modes 50

and their components; guideways, vehicles, control technologies, and institutions. We will examine characteristics of transportation that affect innovation and are common to all modes, characteristics such as the standardization necessary to link individual modes into networks. This analysis will explain the present status of innovation and provide a basis for recommendations to better orient and accelerate innovation and technology deployment activities. Although we will use the extensive literature dealing with the many aspects of innovation processes, our organizing focus is that of indus- try structure, a focus reflected only in bits and pieces in the innova- tion literature. We believe that more attention should be given to the structures of activities that provide the environments for innovation. This atten- tion might clarify the diverse findings of empirical studies such as those reviewed by Johnson (l975, chapter 4).l It also might explain why innovation differs among industries, the factors that condition the diffusion of innovation knowledge and its disregard or adoption, and the social and economic roles of innovative individuals and organiza- tions. Concerns about innovation follow from the heavy investment of government and some industries in research; the regulatory, taxation, patent, and other policies of governments that might accelerate or dampen innovation and its adoption; and the role of innovation in economic growth, including its contribution to the comparative advan- tage of one nation versus others. To respond to these general concerns and our specific interest in transportation, we make three recommenda- tions at the end of this paper: to strengthen the assessment of com- ponent technology development, to better define needs for systems, and to better formulate systems alternatives. Although these recommendations are different from those of studies that have examined the national scene, such as the Charpie (U.S. Depart- ment of Commerce, l967) report, we believe our recommendations have a broad application. In particular, these suggestions with suitable adaption might be generalized to public facilities such as water supply systems, communication systems, and the post office. (Elsewhere, we have written about the rather striking similarities between these sys- tems and transportation [l978]). TRANSPORTATION Viewed in a general, simplified manner, transportation is performed when force is applied to displace a mass (soil erosion, the drilling of cavities in teeth, and the flight of an airplane are all transporta- tion) . Viewed narrowly, a transportation innovation is the organiza- tion of a physical system to perform that work in some purposeful man- ner. Even before the building of pyramids, innovative groups and indi- viduals had thought of ways to enable and control the displacement of masses. Five major transportation modes—rail, air, highway, water, and pipeline—make up today's systems. In order to adopt and deploy 51

technologies, institutional structures were necessary and they were created; railroads and airline companies are pieces of the structure. Some things about transportation industries are not so obvious. Why are there five major modes and not some other number? (Here, we are viewing transit as a variation of rail and highway.) Why do some modes involve both public and private activities and others appear less splintered? Does it matter? Why do innovation and technology activi- ties differ, and what needs to be done to improve those activities? What are innovation opportunities? These are simple questions where simple answers do not give insight. Growth Dynamics One useful way to approach these questions is to think of the modes as evolving in a dynamic of physical systems-institutional systems-market systems. The history of the automobile highway system during this century provides an example. The automobile was the triggering innova- tion, the putting together of the wagon chassis with a steam, electric, or gasoline engine; then came the application of vehicle control pro- tocols to wagons and buggies—the operator guided the vehicle and obey- ed the rules of the road. At first, the automobile was truly a rich man's toy; it was expensive and had little use, for the road system did not accommodate travel. But the situation changed drastically in only a decade or two. By the l920s a paved roadway system suitable for auto- mobile and lightweight trucks was expanding rapidly (Figure l). A variety of innovations such as lightweight steels, improved testing methods, and assembly line production was improving and reducing the cost of automobiles. The market was also adjusting as suburbanization, new patterns of employment, and a different wholesale and retail dis- tribution pattern emerged. The dynamic was energized by improvements in accessibility. The decision to purchase a vehicle enabled the user to gain accessibility provided by the road system and offered by changes in patterns of pro- duction and consumption. The gasoline tax, a financial mechanism, link- ed automobile use to road improvements, but it was truly the expansion of accessibility opportunities that shaped the dynamic. This dynamic involved more than the innovation of a physical system, its deployment, and market adaptions. Institutions were neces- sary; they too required innovation. The Alfred P. Sloan type manufac- turing industry was one such innovation; financial institutions provid- ing installment credit were another. Institutions to provide the high- way system evolved, state highway departments were created or modified, and local government and federal institutions and financing arrangements were formed. Vehicle insurance, driver training, and traffic engineer- ing institutions were also established. 52

-112 1900 1910 1930 1930 WO 1950 I960 !9?0 1980 FIGURE l. Comparison of passenger car sales (sales [Motor Vehicle Manufacturers Association, 1977]H shown every fifth year [l940 to l945 eliminated] to l970, annual thereafter to l976.) with total and surfaced mileage of roads and streets (rural roads and municipal streets [U.S. Bureau of the Cen- sus, l975],5 series begins in l92l, ends in l970). Each transportation technology form has a growth dynamic. Table l characterizes each technology or mode within its dynamic, and the inno- vation and technology adoption that corresponds to the phase of its dynamic. Now we will turn to the reasons for variations in innovation and technology activities and lay a basis for recommendations to improve those activities. The dynamic for a technology has beginning conditions from which it emerges. The highway system provides an example. The King's high- ways of the fifteenth and sixteenth centuries were swaths along which people could walk and drive animals. Improvements were limited to lay- ing stones for footing over poor ground and the providing of narrow bridges. In the l700s, wagon and carriage traffic increased rapidly with the extension of maritime and colonial activities; a dynamic 53

TABLE l. Characterization of Transportation Innovation and Technology Deployment Status of Transportation Technologies and Their Institutions Innovation and Technology Activities Near the end, at, or past their growth dynamic—mass transit, rail freight, automobile Frenetic search for technologies to reduce costs and to meet constraints including: political requirements for service in high cost markets, regulatory, labor, capital, and insti- tutional ; much government involvement in technology matters; technologies of limited scope (e.g., improved ways to empty fare collection boxes, better rail wheels, lightweight automobile hoods); there are narrow (e.g., tech- nology is needed for filling potholes) and sometimes suboptimal views of technology needs; some interest in new systems when the technology is well past its growth dynamic, e.g, personal rapid transit; interest in technologies to protect traditional markets, e.g., TOFC and COFC. In rapid growth phase—truck highway, pipelines, inland waterways, air. Near the beginning of their growth dynamic--slurry pipelines; container, roll-on, roll-off, and large-bulk ships, Alternative technological and/or in- stitutional forms continue develop- ment from early growth dynamic phase, e.g., specialized contract carrier trucks, new aircraft, product and slurry pipelines, and the United Par- cel Service; technology responding to safety and environmental regulation, other constraints may be pushed aside by productivity gains, although they affect the technology, e.g., Air Line Pilots Association work and pay re- quirements; search for technologies for system expansion, e.g., efficient short-range aircraft Search among the technological and institutional forms for old and new markets; high productivity pushes aside constraints other than environ- mental and safety; little government involvement; industry factors seek standardization. 54

responding to demand began. Highways were developed for wagons and carriages, with the assistance of local government and tollway organi- zations. Plank roads were used in the United States. The macadam road in England is remembered from this period, although McAdam's genius lay more in the organization and financing of highway building and mainten- ance than in the type of surface ordinarily associated with his name (Webb and Webb, l9l3)6; there was nothing new about that. Another example of a dynamic running its course, slowing, and be- ginning again, starts with the "break of bulk" steamship in the late 1800s. By the l920s, there was a stagnation of technology (and institu- tion and market) development. Recently, containerization and the use of larger container ships and large bulk ships have set off a new dynamic. The highway system provides several examples of the reenergizing of a dynamic. The interstate system in the l960s enabled higher driv- ing speeds that, together with market shifts, continued to improve access until recently. Early in the century the highway-truck system evolved rapidly, serving mainly a local collector-distributor function. Improvements in the regional roads in the l930s and l940s and develop- ment of the interstate system later set off another truck dynamic which continues running its course. Conditions at the beginning of a development dynamic include insti- tutions and market conditions, each with its claims on resources. Much of the market is subject to the "tooth and claw" of free enterprise; its evolution with the growth dynamic is relatively unfettered compar- ed with other aspects of the dynamic. But market conditions and the difficulties of changing them are not to be dismissed completely. As we have discussed elsewhere (Garrison, l978),•* systems users claim a right to transportation service, and much of government regulation of transportation service and subsidy, such as that of mass transit, is to offset changes resulting from growth and the new interplay of technology forms. Institutional change has a dynamic of its own and usually occurs with the creation of new institutions. Once created, even new institu- tions reflect the conditions of the times in which they were created and become a brake on change. The railroad organizations of today exhibit conditions from the time of their origins. Their geographical division, for example, reflects communications and logistics conditions that existed over a century ago. The railroads have changed, of course, but the basic organizational frame remains. The railroads put carriages on steel wheels; routes were laid out primarily for passenger traffic. In the United States the abundance of coal and early adoption of high-pressure steam engines, along with the constraints on labor, including construction management skills, affected the grades and layout of today's routes (Williams, l976)'. The light- weight four-wheel carriage became a l00-ton or more freight car riding on four-wheel trucks, with up to 36-inch wheels and a much higher center of gravity than the carriage transformed to a railroad car. Although physical technology has changed incrementally and is radically different from what it was in the beginning conditions. G. Plowman has identified (to the author in a letter) the increasing of gauge, doing away with 55

operating railroads on the carriage-wheel spacing of the eighteenth century, as one of the major problems of today's railroads. Later, we will discuss the components of transportation technology-- guideway, vehicle, and control—and how their disjointed character has limited change to building incrementally from conditions existing at the start of a dynamic. Geographical Networks: Standardization The highway system and the waterways system reflect in their present development the existence of an initial network (the then existing road- way and waterway networks). Their present dynamics began with the necessity of serving those networks. Much of the mileage of today's highway system was laid by the late nineteenth century. About 50,000 miles of interstate route have been added, and mileage has been added as cities have expanded, but much of the dynamic growth of the twentieth century has occurred on the stage set by the existing road plant. Today's inland waterway and maritime trade started out with existing routes tied to the locations of ports and markets, and modern inland waterway transportation technology takes place on routes where rafts once floated and steamboats hauled cotton and pork. Railroad, airline, and pipeline networks were new, but even here there is the imprint of preexisting markets and the location of the routes of competing modes from which these new modes hoped to snare traffic. A preexisting network can restrain or assist the innovation pro- cess. It assists because it eases the onus of tying places together by procuring land or terminals. We will return later to the positive side of the right-of-way or network question when stressing innovation opportunities. The sections on transportation system components and incremental decision-making will also deal with the forcing of a tech- nology to operate on an existing guideway. In order to benefit from accessibility, either people or goods have to get from one place to another; there are rewards from connecting links into networks, system articulation, and standardization. A stan- dardized time system was developed, and a standardized railroad gauge was adopted. Other needs for standardization resulted in the creation of the Association of American Railroads and, by other actors, uniform labor rules. Uniform air and highway traffic rules represent standardi- zation in other systems, as does the evolution of pavement construction standards and the development of the rules-of-the-road in ports on in- land waterways. Early, industry seeks standards; the government's role expands later as safety and service standards are demanded. Today, of course, nonindustry-specific government safety and environmental re- straints apply early. Standardization has two chief effects on innovation. First, it almost locks out technological change that is more than incremental; innovations have to fit the standards. It also dampens innovation because of the effort required to meet standards. 56

Yet, standardization forges a large market if a desirable technique or device can be innovated to fit a standard; a large market can result in economies of scale in production. (This seems to be the motive behind creating standards for buses, wheelchair lifts, and similar things.) The critical matter is the ease with which standards can be bypassed. For example, although there are standards for packaging, innovative techniques seem to work within or push aside those standards by obtaining exceptions (U.S. Department of Transportation, l978) . As history shows, standards are pushed aside if a development is highly desired. MoPed advocates were able to sidestep safety standards for motorcycles. The 707 aircraft proved so productive that the stan- dards for runway strength and length were dropped by airport operators who wanted jet service; they lengthened and strengthened their runways. The ascent and descent rate used then in air traffic control suited to DC-3 aircraft and ill-suited for jets was also pushed aside. On the West Coast, the longshoremen constraint gave way to productive container systems. Standards for harbor dredging also gave way rather quickly before the productivity available from large container or bulk ships. Today, the productivity gains to be garnered from increasing truck weights and sizes are clashing with entrenched standards. Market Impacts Turning from standards to markets, we note production and consumption shifts as a dynamic evolves. This market response affects the charac- teristics of innovation during the dynamic. Successful innovations are market-sensitive; attention is given to the manner in which the market is evolving, and more effort is made to fit a technology to new develop- ments or to particular niches in the market. The development in l837 of packet service out of the Port of New York for the North Atlantic trade was responsive to a market niche. Today's specialized ships hauling assembled automobiles are another maritime example. Unit trains and the specialized trucks of contract carriers are other market niche innovations, and efforts to find technologies suitable for short-haul, collector-distributor air transportation represent a sensitivity to the need for fitting technology to market niches as well as an effort to im- prove the network of service. This is the pull of demand. A transportation technology evolves, and production and consumption organizations shift how and where they do things considering the availability of that technology. The continu- ed learning and shifting of the activities or organizations create new opportunities, the growth of organized diversity on the market side. Innovation We can describe several relationships between innovation and technology utilization and systems development. Table l lists in capsule form the activities resulting from these relationships. Below, we outline a development dynamic as it runs its course. 57

0 There is a vigorous, early competition among forms of the tech- nology—vehicle, guideway, and control combinations and their institu- tions. 0 Also early on, managers seek standardization to provide networks to serve markets and to achieve economies of scale in production. o As the market evolves, more and more attention is given to pro- cess and product innovations suitable for particular markets. 0 These technologies for market niches are constrained and limit- ed. As the system develops, existing institutional arrangements and conditions set by the existing physical system limit the scope of tech- nological change. « Attention shifts from the innovation of competitive technology forms to innovations of a very narrow scope, bits and pieces of hard- ware or processes. The impacts of particular technology developments are limited, although their aggregate impact may be great. o While there are restraints on systems from the start, including those common to all activities, transportation-specific system con- straints on innovation and technology development and deployment multiply as the dynamic unfolds. These constraints result partly from increasing recognition of system externalities; they are originated by governments. Many are imposed by the increasing complexity of trans- portation institutions and the inability of complex institutions to overcome stasis. The rights of labor, management, and users are in- creasingly cemented. Capital restraints tighten. o The publics interested in the system become less supportive; they are increasingly disenchanted. At first there is support, for the system dynamic multiplies accessibility. Later, gains are not as great and negative externalities more apparent. The systems' bureaucracies become increasingly inflexible. The public demands innovation and tech- nology, often via regulatory mandates, to fix problems. 0 In contrast, late in the dynamic there are publics who imagine and value early technology and market conditions. They seek maintenance of technology with no market for innovation (cable cars), or a reincar- nation (light [sic] rail transit) using the best available innovation and technology. 0 The concerned publics' and the technologists' views of innova- tion and technology needs shift. Early, they are broadly framed in terms of systems and associated development. Later, they narrow and are addressed to correcting something about a small part of the system perceived to be faulty. 0 The systems' growth is never unbiased; conditions at the start of the dynamic strongly influence its course and the opportunities for innovation. o Yet the path of the dynamic is never certain; the dynamic may be changed fully or partially by market shifts (more need for coal transport) or by technology and institutional development. o The role of the innovator changes as the dynamic unfolds. Early, the innovators focus on the system; later they address bits and pieces of things. Because of the structure of the industry and societal views of needs, the small is valued more than the large. 58

• Yet the above does not restrain innovative effort. Systems innovators do not hold the values of their peers and their institutions. As these values become more rigid, more innovators disclaim, but their probability of success declines. Again, the activities listed in Table l are some outcomes of these relationships; more outcomes could be noted. Not all the relationships are fully discussed, but their basis and consequences will be more fully developed as we continue. Productivity Before developing further the characteristics of transportation that are behind these relationships, we ask whether these relationships matter. In our opinion they do. q In his book on the automobile industry, Abernathy (l978) observed: (l) the development of a dominant technology (in this case, the Model T) and (2) the evolution of a mass production, low-profit- margin method of producing it. He argues that the search for producti- vity is consequently constrained to minor (mass production) process improvements, which are subject to diminishing returns. The future is bleak. In Chapter 4 of his book, Abernathy refers to supporting studies, and in a l975 discussion paper with J.M. Utterback^ (which appears to be the basis for Chapter 4), Abernathy provides examples from the semi- conductor industry, the aircraft industry, light bulb manufacturing, the automobile industry, and the processed foods industry. Here and in his later work he offers a conceptual model in which an industry is created by radical product innovations. One dominant innovation sets a pattern for a product; then process innovations dominate as ways are sought to produce that predominant type. In this perspective of our paper, the restraint on the supplier activities in transportation results from industry structure. Produc- tivity is a question more because of those constraints than because of those of the manufacturing process of particular firms. In contrast to automobiles, railroad cars, barges, and aircraft are produced on a job lot (a run of several) basis, yet they too are productivity limited. Abernathy's view of the firm is too limited; to change productivity re- quires system change. Such change can be achieved if system technology can be innovated and deployed to change a development dynamic or start a new one. Technology and Market Gaps Figures 2 and 3 abstract two performance characteristics from the com- plexity of the organizations providing transportation. Figure 2 illu- strates that the cost of moving a unit decreases as the number of units increases. Figure 3 illustrates that the cost of moving a unit between places, for a given number of units, varies from mode to mode depending 59

on the distance. Other such figures could be drawn illustrating charac teristics of transportation activities. These might display relation- ship of gross to net weight; horsepower per ton moved; velocity, say distance a commodity may be moved overnight; and weight/volume ratios for the filling of vehicles or containers. Data on relationships of this type are displayed in National Transportation Trends and Choices , (U.S. Department of Transportation, l977).nJ. D. Ward et al. (l977) have explored such data fruitfully. cost/ unit Truck volume of shipments FIGURE 2 Relationship of cost to volume. (The dashed lines suggest where existing modes perform best and the "gaps" between them.) cost/ /I unit distance FIGURE 3 Relationship of cost to distance, lines suggest gaps.) (The dashed 60

One may think of mapping between such performance characteristics of transportation, markets, and the technologies and institutions that provide transportation. Mapping on Figures 2 and 3, we would expect rail and water to provide for high-volume, longer-distance movements. Helicopters are used to move structural steel for the construction of towers in isolated places; airlines get a larger share of long-distance traffic than short. Gabriel Bouladon (l967)* has developed a concept of gaps to describe the interstices where the existing modes do not serve markets well: the "too far to walk but too close to drive" gap and the "too far to drive but too close to fly" gaps, for example (Figure 4). DISTANCE (m'les; TIME ImmutM) SPEED (m.p h I eoon 965 1?0 1805 3000 FIGURE 4 Bouladon1s transportation gaps. Bouladon's development of gaps is a useful introduction, but it greatly oversimplifies the manner in which gaps may occur. In passenger trans- portation, gaps should be imagined arising out of the performance characteristics of the available modes compared with the functions that passenger transportation serves and not with distance alone. Elsewhere, this author and Clarke (l977)l4 have sketched the concept of a neighbor- hood car that, although serving in the range of "too far to walk but too close to drive," fills the gap defined by those functions that are performed in neighborhood travel. The camper vehicle fills a certain kind of gap. When sketching gaps for freight transportation, one should also consider functions as well as the comparative advantage of modes, a complex mapping. Institutions compete with each other and strive to preserve them- selves. The interstices between which existing modes have a marked 6l

comparative advantage are competitive battlegrounds. The railroads fight hard to maintain their claim on bulk traffic vis a vis the agricultural- exempt truck and waterways. Some railroads give attention to trailers and containers on freight cars (TOFC and COFC) and "merchandise" traffic in order to preserve their traffic or expand into that gap. Urban bus operators are preoccupied with expanding service in thin markets to com- pete with the automobile, and much of what is said to be needed in tech- nology development of mass transit is technology to serve those markets, for example, paratransit. Questions can be asked (but not answered) about the appropriate- ness of the present number of modes. For example, in terms of existing markets, is Amtrak needed given intercity bus service? The extent of railroad freight service is sometimes questioned given truck service. Is a complex of long-distance highways engineered for automobiles reasonable given the availability of air transportation? Intermodal arrangements combine the advantages of two or more modes. Technologies to improve intermodal service were stressed years ago in the Eastman Report (Office of the Federal Coordinator of Trans- portation, l940),^ but these are not of interest to the existing modes when one party has something to lose. (It is not surprising that rail dominated services [ICC Plan II] are the more successful TOFC endeavors.) An intermodal service works if one party is not harmed and the other gains, or if both gain. Rail transit institutions will pave parking lots and worry about bus stops adjacent to stations. Maritime container shipping organizations and railroads are jointly concerned about effi- cient ports and intermodal service; those who carry containers to the port do not have the option of continuing overseas with them. We will return to this notion of gaps and develop innovation and technology options more fully after discussing some of the character- istics of modal components. COMPONENT DISJOINTEDNESS AND INCREMENTALISM Physical work in transportation is performed by applying a force to move something along a guideway. While the guideway may control the direction of movement, there is always additional control activity. The core of the physical technology involves a vehicle and propulsion unit, a guideway, and a control system—the components of transportation tech- nology. There are technological and institutional supply streams for each component. Highway vehicles, for example, are provided by automobile and truck manufacturers through a system involving dealerships, financ- ing institutions, regulating institutions, fuel suppliers, and so on. The technologies here are mainly those of mechanical engineering for the vehicle (and chemical engineering for fuels). Guideways are supplied by governments drawing on several subspecialities in civil engineering. In the main, control of the vehicle movement is provided by the driver training and licensing systems. Traffic control is provided by traffic engineers, and the driver's fiscal integrity is warranted by insurance arrangements. 62

This pattern of disjointed technology components appears in each transportation mode. Ships are provided by shipbuilding companies, operated and controlled by shipping companies and by inshore and offshore rules-of-the-road, and operate on guideways partly provided by naviga- tion and by dredging. Aircraft are produced by airframe manufacturers, operated and controlled by airline companies and by air traffic control regulations, and use guideways supplied by governments. One outcome of this disjointedness is incremental decision making. The technology-supplying institutions see their role as that of meeting needs occasioned by markets and constrained by characteristics of other components. Railroad right-of-way suppliers strive to provide suitable guideways for the types of trains that are operated on routes and their frequency. Automobile manufacturers have an eye to the market. They are constrained by the type of highways on which the vehicles will be used, the standards of driver licensing and the norm and distribution of driving skills, and traffic rules. Highway traffic engineers establish regulations considering vehicle characteristics, drivers, and roadway conditions. The consequences of incrementalism are constraints on the tech- nology supply stream. Only incremental technology change is permissable, and all technology choices that consider a system—that is, involve con- trol, vehicles, and guideways—have no market. This can be seen by even a cursory examination of current technology activities. The traffic engineering literature is replete with ways to do traffic en- gineering better; all else is given. The protocol for benefit-cost studies in highway design is to minimize the joint cost of providing highways, given the characteristics of vehicles including their operat- ing cost and the way they are operated. Today, automobile manufacturers are preoccupied with developing technologies to meet emission and fuel consumption standards with everything else taken as given. The auto- mobile engineering literature is as limited to the automobile (and the truck) as traffic engineering literature is limited to traffic. An interest in transportation systems is mainly a conceptualiza- tion of transportation as a network rather than a link and node phenomenon. Systems planning, such as that of the United States Rail- way Association (USRA), is planning for guideway systems; urban trans- portation planning since World War II has been planning for highway networks. Symptomatic of the lack of system thinking is the recent change of name of the Institute of Traffic Engineers to the Institute of Transportation Engineers—traffic engineers take what they do to be transportation engineering and do not seem to recognize the system scope of the technology. Finally, consider studies of highway needs such as the l977 study published by the Congress (Committee on Public Works and Transportation, l977).l6 For many years, studies of needs were made by comparing the physical state of existing highways with an ideal expressed by engineer- ing standards. In recent years there has been a modest recognition of markets through considerations of the amount of traffic on facilities. Highway needs studies take other components of the system as given. In addition to being highly constraining, the disjointedness of components distorts system goals and innovation. Technologies are 63

sought without institutions to conceptualize transportation and thus formulate broad transportation goals. Goals, such as good roads or safe cars, reflect institutions and components. Transportation is heavily regulated, which also affects goals. Some institutions strive to meet regulations regardless of the cost. Compared to other modes, railroads appear to have integrated components. Railroad organizations supply their own guideways, make decisions to purchase equipment, and control and operate that equipment. Even so, these components are supplied and operated in a disjointed fashion largely because of the separate technological traditions of components and their supplying institutions. Component managers have differing goals. The equipment manager seeks appropriate equipment, given operations and guideway systems; the guideway manager seeks an appropriate guideway, given equipment and the way it is used; and opera- tions personnel spot cars and move them, working with what is available in the way of equipment and guideways. The disjointedness of railroad components is illustrated by the aggravation of right-of-way problems through the purchase and use of heavy freight cars. Cars of l00 tons or more created unexpected right- of-way and operation problems, leading John C. German of the Missouri Pacific Railroad to remark, "there has not been enough cooperative discussion between the equipment engineer and the track engineer" (German, l974).l7 Railroads are heavily regulated—self-regulated through industry- wide standards and regulated by government. Industry standards apply to components, and federal regulations apply mainly to operations deal- ing with service and rates. Safety regulations are addressed to com- ponents . The other modes are also regulated. Gellman (l97l)l8 has stated that regulation has distorted decisions to purchase equipment and has dampened equipment innovation. He concludes that "the innovative per- formance of the transportation sector can best be improved by gradually eliminating economic regulations." While we accept Gellman's remark about distorted and dampened innovation, we do not agree fully with his conclusion. Even without regulation, decisions to purchase equipment would be constrained; they would be incremental and oriented to com- ponent goals. Innovation and its adoption are often motivated by an opportunity for a specific payoff. A.S. Lang and S.A. Burd, (American Association of Railroads, l976). in an unpublished paper responding to suggestions by Wyckoff (l974),19 Wyckoff (l976),20 and Reebie and Robertson (l979),2l have addressed the difficulty of forming profit centers in railroad organizations. Profitability is a matter for the chief executive officer because expenditure and cost control are in the hands of equipment and guideway providers and operations staff while revenues are in the hands of persons recruiting traffic. Lang and Burd examine options for align- ing cost and revenues on a more decentralized basis, either through sub- organizations that recognize specific lines of business or by spatial markets. But equipment and routes may be used for more than one line of business, and spatial markets are not discrete. A particular spatial market, such as the city pair, is entered and egressed by traffic serv- ing other markets. To deal with these complications, Lang and Burd 64

suggest organizational forms where equipment and operating people func- tion as profit centers selling to marketing people. Lang and Burd's suggestions offer insight and should be useful. We do not, however, feel that these organizational reforms would deal with the fundamental problems of incrementalism and disjointedness. These problems will remain because component supply streams are en- trenched, their goals are internalized, and constraints are imposed by network articulation and standardization. Lang and Burd observe that less-than-a-truckload-lot intercity trucking firms have been able to organize profit centers for a particular market (cities) and contract truck carriers organize to serve specific markets, often using tailored equipment. While this capability contributes to efficiency in the firms organized in this manner, it is very limited, being constrained by com- ponent disjointedness. Equipment and operations decisions are made incrementally against the backdrop of conditions of the existing high- way system and traffic control with its speed and weight controls. IMPROVING COMPONENT INNOVATIONS Useful recommendations for improving innovation, technology development, and deployment within extant component supply arrangements are limited. Institutions, policies, and programs exist in the planning, managing, and expending of resources, but the restructuring of institutions is slow and difficult. Restructuring also is not practicable because com- ponent-arranged activities reflect the professional and scientific disciplines and the organization of the trades; relationships are cemented into existing status and behavioral patterns. Transportation organization is a slice of all social and economic organization, and change driven by transportation has little priority and voice. For this reason, we regard present arrangements as fixed and seek actions consistent with those arrangements. We seek action that will better match component activities to systems needs and opportunities. We seek to lessen the risk that component technology may suboptimize. Suboptimization Component-shaped innovation and technology adoption is quite active: improved electronics for aircraft, electronic engine control for auto- mobiles, computer-aided vehicle design, improved insulation for tank cars, improved aggregates for highways, better methods to preserve rail- road cross ties, and active or passive sensors for traffic control. This innovation within components is viable because it passes the market test of being useful to the component. Its efficacy for systems is un- known. We believe that component goals are not consistent with systems goals that are almost never stated. The ability of systems to perform social functions measured against resource use may or may not be im- proved by a particular component innovation. A possible example is the 65

procurement in recent years of heavyweight railcars with large wheels by the mechanical departments of railroads. These perhaps have not been cost-effective if their cost is extended to the damage that they do to rail guideways. Earlier, we commented that the development of the 707 aircraft violated the constraints of incrementalism because it (and the family of jet aircraft emulating the 707) required the strengthening and ex- pansion of runways, changes in aircraft control protocol, and changes in terminals. Then, many thought that the 707 did not fit the system, especially its market interrelations (Bright, l978).22 Yet, in this case system change occurred because of a major change in a component. To some extent the diesel railroad locomotive and the development of the interstate both have induced system change. While these sytem changes probably measure positively in social and economic terms, an unanswered question is whether some other technological form might have been created if overall system impacts had been considered in the beginning. Assessments There are existing mechanisms addressing the worth of new technology, programs, or projects: technology assessment, environmental impact, and inflation impact analyses. Inflation assessment is concerned primarily with the trade-off between a proposed action and productivity. Regulatory actions are reviewed, actions that may induce hardware inno- vation or the adoption of the existing innovation. Technology assessment has been institutionalized in the Congress, in federal agencies, and in some state and local governments. It takes the stance of measuring technology impacts on scales of efficiency, environment, energy, and externality. The recent study of the automobile by the Office of Technology Assessment of the Congress (l979)" Was one of the most thorough assess- ments made of a transportation technology. The study examined where the technology is, how it might evolve, and its associated costs; it treat- ed air pollution and fuel problems. The cost of implementing more fuel- efficient and environmentally benign vehicles was assessed. Market and systems matters were not much recognized. The study was a vehicle assessment and never claimed to be more. It dealt with mobility, acces- sibility, and related topics of the growth dynamic in only a very slim fashion. It took the way things are going as a given, including the present and future status of the nonautomotive components of the system. Most technology assessments in transportation take a similar stance, although they are generally less broad and thorough. Environmental impact analyses and statements are mandated for all federal activities with "significant" impacts on the environment; they are also used by a number of state and local governments. While the mandate is broad, the emphasis is on physical environment impacts. Analyses are most often addressed to designs and plans using existing technologies, projects ready to be implemented. Environmental impacts are listed and quantified when practicable; decision makers may thus 66

incorporate these assessments in decisions about whether a project will go forward and, if it goes forward, about what steps should be taken to mitigate environmental impacts. Because the environmental impact state- ment is addressed most often to projects—a railroad relocation around the town or a highway alignment and interchange—it considers the triad of components. An airport impact analysis for example, might examine noise alternatives considering the type of aircraft to be used, approach and departure control procedures, and runway and apron alignments. These regulatory, technology, and environmental assessments have the appearance of being a choke between innovation and subsequent infor- mation dispersion and technology adoption. However, the existence of assessment activities affects the way programs are devised and evaluat- ed well before the hurdle of the assessment occurs. This thinking ahead is a feature of technology implementation planning (TIP) suggested by House and Jones (l977). In this approach, the problems of implementa- tion are thought through and fed back to the program design, the evalua- tion of its milestones, and the selection of program options. A Suggestion A modest policy suggestion is that all component technologies be assess- ed in light of transportation systems and market considerations. This could be done as an addition to current technology assessment and en- vironment impact assessment or both. A strength and weakness of this suggestion is that it is like the requirements of present policies. This is a strength because actors and institutions might be comfortable with the suggestion. But to avoid not doing more than is now being done, it would be necessary to spell out what is meant by components and systems and market evaluation. Present-day market evaluations are static; they will, say, compare rapid transit with the automobile and ask how many riders will be diverted. It is important to state dynamic questions and inquire into market ad- justments and development paths for a technological system, its insti- tutions, and its markets. As discussed, most of present-day "system" analysis is limited. To improve the limited sense of system, programmatic guidelines should spell out interest in how the parts of the technology and their insti- tutions interact with each other, and again, development dynamics should be highlighted. This modest suggestion should have more than modest results. An example of a response to a call, such as that by Carey (l979), for a new round of research, innovation, and technology deployment for extant highway bridge types will make the point. Many urban and rural bridges have reached the end of their useful life. Many were bult in the l920s as the road system was upgraded. Even newer bridges have had their life shortened by increasing loads and the use of salt to melt snow and ice. Inventories of repair, re- building, and reconstruction needs indicate that they are massive. The funds are sought from Congress for a bridge program. 67

But a consideration of the system and how it is used suggests that innovation of new technology is misguided if it is to support the re- construction, rebuilding, or replacement of bridges in the places that they are now and with their present traffic loads. "The bridges are falling down, build them back" is an incorrect view of the problem. The rural highway system (except for the interstate) and the older parts of the urban parts of the system were laid out in the horse, buggy, and wagon days. The over-the-road cost of transportation was high, while the cost of the construction at that time was relatively low; except in the downtown of cities, traffic was light. The situation is radically different today. The relationship between variable (vehicle) and fixed (guideway) costs has changed greatly, and increased volumes of traffic offer opportunities for economies from the concentration of traffic on routes. Consequently, a system for today's markets would have many less miles of route. Some routes should be built with higher standards than today's to carry heavy traffic and to gain economies for the vehicles and the routes. There would be more circuitry of travel. Should those bridges be rebuilt in the pattern that was right for the early part of the century, or should innovation and technologies be sought suited to routes where economies of scale are achieved in heavy traffic volumes and in heavy loads? SYSTEMS INNOVATIONS Marked improvements in transportation productivity and the creation of options for social and economic development depend on the innovation of transportation systems, though not exclusively. Innovation is needed now for energy shortfalls, and rapid increases in the cost of energy and other resources will affect the evolution of social and economic organizations. As these organizations adjust to new conditions, it would be highly desirable to have a variety of supporting transportation op- tions. Further, the options provided by the present characteristics of transportation will be even more restrictive as higher fuel costs im- pinge on the performance of transportation, a factor also pressing for major improvements in old systems or for new systems. Working toward suggestions with respect to recognizing and creating new systems options, we shall return to two matters that have been dis- cussed previously: the potential for using existing guideways and tech- nology gaps. We shall then review some examples of recent proposals for system technology and use these to pinpoint what seems to be needed. Two suggestions follow. Using Existing Guideways As noted, a transportation system may be created de_ novo or developed by revitalization of an old system, but in either case it is tied to preexisting market conditions and route structures. The liabilities of preexisting route structures have been mentioned, but preexisting routes can be an advantage. 68

The argument is simple. There is excess capacity throughout all systems because of high over-the-road cost relative to guideway cost during the early development of modes. Earlier, dense highway and air networks were used. Ports and maritime routes were everywhere that ships could sail, and railroad routes were more ubiquitous than today. For a variety of reasons including important political ones, much of the excess capacity created by shifts in cost has not been abandoned; it exists and provides spaces for guideway development. Many of the problems of acquiring land and expanding capital are mitigated. The idea that there is little or no excess capacity is incorrect. There are some urban airports, air space, and urban roads and streets that are congested a few hours every day. This congestion is visible and annoying, but it obscures the fact that these facilities are not busy most of the time and that they are only a small part of the whole. (We do not view this congestion as serious. It would be ridiculous to construct facilities so that they were never to be used at capacity [although most of the guideways for the modes are so constructed], and there are available tools, such as pricing, that could improve conges- tion management.) Gaps are difficult to identify because of the equilibrium between the transportation available and the organization of social and economic activitity. However, social and economic organization responds to many factors other than transportation, so the relationship between it and the availability of transportation is less than perfect. This suggests that "market pull" gaps may be identified by continually monitoring and evaluating social and economic change. The existing modes are continu- ally searching for markets, and market response from new services suggests another way gaps may be identified. Finally, more thorough studies of how the transportation system performs or might perform "technology push" would improve understanding of situations where im- provements in performance might provide a market for innovation and technologies. Examples of Systems Technology Examples of proposed systems technologies will illustrate proposals and point out what is needed in order to better generate options. (See also Gabor, l970.)26 0 Railroads haul coal on unit trains between points of production and consumption using old technology. Given the quantities involved, available vehicle and control technologies, and the availability of guideways, it might be reasonable to operate vehicles with, say, a l6- foot width, with a self-contained propulsion unit or units, say, elec- tric motors on each wheel, with either offboard pickup of electric power or a generator on the car or in a locomotive. To keep down rolling 69

resistance yet avoid high pressure where the vehicle touches the ground, large rubber tires might be used on ribbons of pavement. Wire-following control would be practicable. Because of numerous grade crossings and to keep down the forces working on vehicles, relatively low speeds would be in order. Present guideways would be suitable except for earth- work requirements; some bridges may have to be rebuilt. Because these vehicles and guideways would be in special service, they would not have to be standardized to other rail equipment, although equipment suppliers might push standardization. Present tracks could be left in place and the guideway used for conventional trains. o The Advanced Freight System study undertaken at the Transporta- tion Systems Center (l977)^7 proposed and examined a TRAILS technology— a l20-mile-per-hour, steel wheel on rail vehicle for COFC, which would share guideway with the interstate system and operate under electronic control. The cost effectiveness of this system versus truck transpor- tation was found marginal. 0 Automobiles and light trucks came first. Although lane widths, pavement strengths, and bridge strengths have been increased and some grades have been reduced to accomodate trucks, the highway system is mainly for automobiles. Development of truck-only routes, taking advan- tage of the economies of higher weight as identified by Winfrey et al. (l968),28 might be practicable. o Proposals have been made for transmission of suspended solids in pipelines, such as coal slurry, capsule in pipeline systems, automa- ted personal (group) rapid transit, and automated highways. o Several years ago there was interest in (passenger) high-speed ground transportation utilizing magnetic or air levitation and linear induction motors. High-speed ground transportation systems would be used to link major cities in the 50- to 200-mile range. Observations may be made about our proposals that might be general to all systems proposals. Interest in systems technologies is sustain- ed when their delivery would utilize existing component technology supply streams—automated highways and transmission of suspended solids in pipelines, for example. Most proposals embody technology develop- ment external to the transportation system, control technology for example. Some proposals are responsive to increasing magnitudes of freight shipments or passenger travel in existing markets: high-speed ground transportation and the TRAILS system, for example. For the most part, these proposals ignore the energy problem, changing social struc- ture and patterns of work and leisure, and changes in manufacturing and distribution systems. Two things are missing: (l) an understanding of the universe of all possible technologies and whether these examples are sensible re- presentations of possibilities,and (2) an understanding of the functions of these examples in terms of market dynamics. Our recommendations will respond to these points. 70

Understanding What is Needed James Hillman (l979)29 recently published an essay titled, "Psycholo- gical Fantasies in Transportation Problems" in which he examines the transportation expert's statement of transportation problems in terms of efficiency and equity. In his analysis, he probes the expert's com- plaints about car repair bills, potholes, and congestion and concludes that the expert has a personalized view of the city and of spatial organization. Hillman's thesis relates these perceptions of transporta- tion, of the city, and of spatial organization to the expert's state- ment of transportation problems. Congress and lay people are accused of knee-jerk reactions to transportation experiences and of demanding naive programs to meet needs. Hillman suggests that the expert's views are also based on fantasy. The initiation of thorough studies of transportation, where needs are embedded in actual and possible directions of social and economic development, would respond to such limited sense of problems and improve recognition of needs for innovation, technology development, and imple- mentation programs. While there are projections of freight traffic and passenger movements based on the way things are now, there is no work viewing transportation as an integral part of social and economic dynamics. (Studies, such as those of the National Transportation Policy Study Commission (l979)^0 using national economic models, are rooted in input/output matrices that take technology and the structure of production as given. They ask about changes given a changing mix of outputs and related price changes, a very limited question from social, technology, and industry structure points of view.) Some properties of the several examples of recent proposals for transportation technologies were noted, including their very limited view of the market for transportation. For example, the TRAILS study compared the cost of TRAILS movements versus truck and rail alternatives. But shipment using TRAILS would double the radii for overnight deliveries and more than double the geographical size of distribution or market areas. This increase could have radical effects on the patterns of manufacturing and distribution; it would extend the market available for daily resupply in a very significant way. In l940, Norman Bel Geddes-^l published Magic Motorways, a simply written and well-illustrated book about transforming the highway system through interstate type facilities, automated car control, and other technologies. There was a chapter on the elimination of graft (that is, institutional improvements), another on opportunities for better fitting transportation into the physical environment, and the book was dedica- ted to the "generation of our grandchildren to whom all that is written here will be commonplace." Bel Geddes' work was imaginative about what society could do with technology although he worked during the l930s, a time beset with problems. Work as imaginative as Bel Geddes' is needed in the l970s, l980s, and beyond to outline what transportation can do. A group should be charged to do such work. 7l

A group that would develop understanding of transportation needs and opportunities cannot be located in existing component supply streams or even close to those supply streams. In government, existing modal agencies and their commitments to components would place political con- straints on the work of a group attempting to understand transportation needs and opportunities. Congressional advocates aligned with modal agencies would also influence the work; consequently, if the work is done in government, it would have to be done externally to the DOT and other agencies committed to existing transportation activities. Congress has supported transportation policy studies on a regular but intermittent basis for several decades. If the Congress could be persuaded that a continuing inquiry into the needs for transportation was warranted in order to make those policy studies effective and in order to improve its cognizance of executive branch programs, then a continuing study that was congressionally housed and supported might be workable. The Office of Technology Assessment would not be a proper home for the needs study; it responds to special needs, and its commit- ment to transportation seems limited. (There have been recent reduc- tions of transportation professionals (Smith, l979). ) The Transportation Systems Center (TSC) in Cambridge might be delegated the mission of performing needs studies, but at this time the TSC is higly beholden to the modal agencies, so independent studies might not be practicable. Understanding Technology Possibilities Again, it must be stressed that it is not enough just to study needs. These would have to be continually transformed into an imaginative set of options for social and economic change, and they would have to be continually integrated with a systems technology development program. This transportation system development activity could be lodged (expanded) in the Research and Special Programs Administration of the DOT. The modal administrations may attempt to restrain systems work to things more closely related to their component technologies. That liability, however, is offset by the asset of their knowledge, which, while applied to components of their interest, might be transferred to new systems. Also, the feedback of systems ideas to component managers might have value. The systems technology development activity would depend politically on the argument that technologies under development are supporting the options being identified by the continuing needs study, so there would be a powerful extradepartmental force influencing the work of the systems group. Continuing attention to developing understanding of transporta- tion performance and technology gaps would provide directions for the systems development group as would the requirement for information by the component technology assessment activity discussed earlier. 72

CLOSURE Our recommendations on improving component technology and systems are addressed to the federal government because of national interests in vproductivity and development. Others have roles and opportunities— the private sector, the universities, and state and local governments. Robert Fulton invented the steamboat; Robert R. Livingston provided financing; state government helped by granting Fulton a franchise to provide service on the Hudson River. Regardless of the national interest, transportation must respond to locally expressed goals and the availability of resources. Further, there is little reason to believe that the locus of creativity is the federal government. There are roles for all creative actors and institutions, and government should configure their activities to encourage and support these roles, REFERENCES l. Johnson, P.S. The Economics of Invention and Innovation. London: Martin Robinson, l975. 2. U.S. Department of Commerce. Technological Innovation: Its Environ- ment and Management. Washington, D.C., l967. 3. Garrison, W.L. "Thinking About Public Facility Systems." In The National Research Council in l978, pp. 244-265. Washington, D.C.: National Academy of Sciences, l978. 4. Motor Vehicle Manufacturers Association. Motor Vehicle Facts and Figures '77. Detroit, Mich., l977. 5. U.S. Bureau of the Census. Historical Statistics of the United States. Washington, D.C., l975. 6. Webb, S. and B. Webb. English Local Government: The Story of the King's Highway. London: Longmans, Green and Co., l93l. 7. Williams, E.W., Jr. "Government Transportation Policy and Techno- logical Progress." Transportation Journal, l6, 86-9l, l976. 8. U.S. Department of Transportation. 8th Annual Report (on) Hazard- ous Materials Transportation. DOT/RSPA/MTB-78-3. Washington, D.C., l978. 73

9. Abernathy, W.J. The Productivity Dilemma. Baltimore: Johns Hopkins University Press, l978. l0. Abernathy, W.J. and J.M. Utterback. "Innovation and the Evolving Structure of the Firm." Working Paper, HBS 75-l8. Harvard Univer- sity, Graduate School of Business Administration, Cambridge, Mass., l975. ll. U.S. Department of Transportation. National Transportation Trends and Choices. Office of the Secretary. Washington, D.C., l977. l2. Ward, J.D., et al. Toward 2000: Opportunities in Transportation Evolution. DOT-TST-77-l9. Washington, D.C.: U.S. Department of Transportation, l977. l3. Bouladon, G. "The Transportation Gaps." Science Journal, 3_, 4l-46, l967. l4. Garrison, W.L., and J.F. Clarke, Jr. Prospects for Neighborhood Cars. Berkeley: Institute of Transportation Studies, University of California, l977. l5. Office of the Federal Coordinator of Transportation. Public Aids to Transportation. 4 vols. Washington, D.C., l940. l6. U.S. Congress, Committee on Public Works and Transportation. The Status of The Nation's Highways: Conditions and Performance. Washington, D.C., l977. 17. German, J.G. "A Maintenance Officer's View on the Effects of Freight Car Dynamics." In Proceedings, llth Annual Railroad En- gineering Conference, p. 49. Washington, D.C.: U.S. Department of Transportation, Federal Railroad Administration, l974. l8. Gellman, A.J. "Surface Freight Transportation." In Technological Change in Regulated Industries, edited by William M. Capron, Washington, D.C.: The Brookings Institution, l97l. l9. Wyckoff, D.D. Organizational Formality and Performance in the Motor-Carrier Industry. Lexington, Mass.: Lexington Books, l974. 20. Wyckoff, D.D. Railroad Management. Lexington, Mass.: Lexington Books, l976. 2l. Reebie, R.S., and A.C. Robertson. Organization of Needed Programs Within the National Rail Transportation System. FRA-OPPD-79-3. Washington, D.C.: U.S. Department of Transportation, l979. 22. Bright, C.D. The Jet Makers. Lawrence, Kansas: The Regents Press, l978. 74

23. U.S. Congress, Office of Technology Assessment. Technology Assess- ment of Changes in the Future Use and Characteristics of the Auto- mobile Transportation System. Washington, D.C., l979. 24. House, P.W., and D.W. Jones, Jr. Getting It off the Shelf. Boulder, Colo.: Westview Press, l977. 25. Carey, W.N., Jr. "Transportation Engineering." In Workshop to Define Needed Civil Engineering Research in the United States. National Research Council, Washington, D.C.: Transportation Research Board, l979. 26. Gabor, D. Innovations: Scientific, Technological, and Social. London: Oxford University Press, l970. 27. U.S. Department of Transportation, Transportation Systems Center. Advanced Freight Systems Study: Phase I. OS 868/R85l5. Cambridge, Mass., l977. " " ' " 28. Winfrey, R. et al. Economics of the Maximum Limits of Motor Vehi- cle Dimensions and Weights, vol. I. FHWA-RD-73-69. Washington, D.C.: U.S. Department of Transportation, l968. 29. Hillman, J. "Psychological Fantasies in Transportation Problems." University of Dallas, Irving, Tex., l979. 30. National Transportation Policy Study Commission. National Trans- portation Policies Through the Year 2000. Washington, D.C., l979. 3l. Bel Geddes, N. Magic Motorways. New York: Random House, l940. 32. Smith, R. "Thermidore at OTA." Science, 205, 570, l979. DISCUSSANT'S COMMENTS BY DORN C. McGRATH, JR. A discussant of a paper can do two things. One is to go through the paper point by point, and suggest gaps and highlights and elaborate on points expressed but not emphasized in all perspectives by the author. Another is simply to elaborate on some of the themes and some of the ideas that the paper provides. I have chosen to do both, inasmuch as it seems to me that the paper deals a lot less with what was programmed as the "Setting for Innovation in Transportation," than it does with some of the processes by which transportation technology has evolved. In the context of the other panels and the other papers, some discussion of the setting qua "setting" is mandatory for us as a panel in any event. In the planning session at Woods Hole in August l979 75

during a day's participation by Bill Garrison, author of the paper, we were not able to cover all points so we did later elaborate on some other ideas that seemed important then as part of the setting. I would like to look at parts of the paper, because there is included one impor- tant idea that "gaps" and "niches" in the transportation services being performed have been stimuli for the formation of new technologies and some administrative systems. Examples range from the packet boats of the early nineteenth century to the perceived gap in the ability of the U.S. postal system to deliver small parcels and goods of high value quickly to other places. Some other niches emerge as opportunities for specialized trucks for short-haul activity or for intermodal transfer cranes for handling containerized goods. All of these have emerged in response to important symptoms of weakness in the original transport systems, and they reflect the fact of new market opportunities that seem to have inspired technological change and, upon occasion, bona fide innovation. Not every author would have done this, but Bill Garrison goes on to seek to fill many of these gaps with recommendations and proposals. In effect, he has accepted the challenge of responding to the mandate of most of the other panels, too, and offers some actual suggestions about how applied technology can fill some of these gaps. But there are two basic aspects that the paper seems to present. One is that clear defi- nition of the fruitful opportunities arising because of the gaps and niches in the service, that we think of as transportation, helps to produce technological change—change ranging from the container terminal as a whole to the roll-on, roll-off systems that break down some of the barriers to intermodal transfers and ultimately lead to new kinds of investment in equipment reflecting a new order of magnitude in weight- handling. There is a question as to how far that trend can go before we should slow the investments in response to the new dynamic that the paper suggests—the dynamic set up by the use of large container ships and bulk-cargo ships, which in turn require new kinds of materials- handling equipment, whether for containers or for bulk cargo. These are analogous to some of the new kinds of materials-handling equipment for transcontinental shipment of coal in slurry pipelines, and mechanical conveyors or other systems also developed to fill gaps. The author maintains that thus a new dynamic is set in motion. I would note that a characteristic of the current context of the setting for innovation is that this dynamic is often undisciplined. We see that it may, par- ticularly in the case of the container ship terminal, become a force motivating every local business promotion group to have its own con- tainer terminal whenever possible, whether it be in Charleston, Savannah, Norfolk, Baltimore, or Boston and New York, for example. Every port suddenly believes in the necessity to compete in that particular field of cargo handling, whether or not the market realities of their port system and their hinterland would make this reasonable. They feel com- pelled to invest. There are many who are prepared to help them do that, and to sell them equipment, outside of any organized context of market reality. There are some recently developed pier projects going begging in New York. There is chronically much unused terminal container capa- city in Boston and in other places that we will undoubtedly hear more about. These are some of the consequences of the unfettered dynamic. 76

There is also a note in the paper that the future looks bleak with respect to productivity. I think that requires another whole set of judgments that we may not be able to make adequately in the context of this paper. But nonetheless, I am not sure that it does look bleak. We were reminded in one of the previous papers that times were, never more bleak than during colonization, when all kinds of innovations indeed occurred. I will not dwell on that because I do not want it to be taken out of proportion to some of the other comments that the paper contains. The paper mentions under "Improving Component Innovations" in the section titled, "A Suggestion," that it is important to state dynamic questions and inquire into market adjustments and development paths for a technological system, its institutions, and its markets. One thing that this paper on the "Setting" does not bring out is the institutional context in which innovation occurs. I want to come back to that, because as a theme it suffers in comparison to the dominance of interesting technological issues presented throughout. This leads to some other conclusions that I would offer as elaborations of the basic points made in the author's paper. It seems that the government, as part of the setting, is seeking to provide some form of organized reconciliation of a series of technological and social factors including human well-being and that these factors become combined in any number of issues involv- ing transportation. The institutional setting as we know it is quite chaotic. The federal agencies are most unpredictable about focusing on long-term goals and objectives. I was very encouraged by the sugges- tion offered by Charpie and Goldmuntz that government ought to stay out of the direct process of innovating and should instead catalyze innova- tion by others. Both also suggest that perhaps a contest sponsored by government with the objective of producing innovations would bear fruit. There are some serious problems, however, with the contest that the government already runs every year. It is called a Request for Proposal (RFP) contest, wherein people try to get funds to do whatever they may be able to get funds to do—not necessarily what is needed, nor what is necessarily innovative, but funds for whatever the government seems to be prepared to sponsor for a short period of time. The rules of that game change, except that the RFP process, one of the more stultifying modes for conducting the game, lingers on. The system has many faults, but one of the worst is that it tends to prevent the giving of free rein to innovators. It also tends to work against giving new ideas time to be developed and to mature. Such ideas may require more than a single budget year. Another characteristic of the governmental side of the setting is the clear lack of institutional memory; each time the administration changes, seemingly, all recollection of anything that went before, ex- cept the integration of railroads, is eclipsed. Suddenly, new initia- tives have to be asserted, even though much of the research that led up to that point might have been quite good, and brand new people bring to bear old techniques in merchandising programs using the progressively imperfect RFP system. Typically, this context of governmental maneuver- ing in place lasts for about two and a half years and then it becomes 77

apparent that something else needs to be done. As election time gets closer, the tendency to innovate falls off very rapidly, as does the discipline concerning where the funds will go and for what purpose. This is part of that chaotic institutional setting that I do not believe should be ignored. It is difficult for people who want better service and ideas about possibilities for improvement out of the transportation system to relate to that process. The kind of contest that the Atlantic Richfield Oil Company offered in l974 had some interesting dimensions. They thought that for a six- month promotional effort, which they freely admitted was to help sell gasoline, they would promote a little contest on ideas for innovation in mass transportation. Some of you have seen the little booklets that were produced to publish the results. It turned out that two years later they had received 24,000 entries, 5,000 of which were from children. All kinds of people were interested in the idea of getting from here to there in different ways. While helium-filled bicycle tires to ease the effort in going up hills may have sounded capricious to us, to a kid, and to others who see something really fascinating about trans- portation and mobility, it was creative thinking. It is what we all believe in very much. But some kind of a contest to tap that resource of frustration and creativity in our market-oriented society might make a lot of sense, if it were organized a little differently. The pent-up creativity and frustration about transportation are certainly part of the context in which things ought to be placed. Also part of the con- text is the reality that we now have a fairly turgid, immobile, and highly organized system of research with an inertia all its own. Also the lack of a sense of what we are actually trying to do is a problem of the institutional setting for innovation. People are filling some interesting gaps in rail transportation by inventing bigger cars, wider rail gauge, and more powerful locomotives. Such approaches are general- ly for one purpose—to take coal from the mine mouth to the generating station or steel mill. Some interesting additional hardware can be generated that way, but many other more basic problems are ignored in the process. In terms of what transportation does for society, for the city, or for the people, in terms of more than one mode, we have not been very purposeful with our innovations. That is one of the gaps in the present setting. A curious analogy comes to mind. It seems to me that if we were as serious about dealing with the problems of domestic transportation as we are about the transportation of armaments and ex- plosives overseas by means of high-trajectory vehicles, to land with extreme precision almost anywhere in the world, we would be much further along. In the MX mobile missile system that is being installed in one large area, with a highly sophisticated set of transporters and controls in order to locate weapons at various unpredictable stations, we have developed a kind of shell game analogous to trying to find a bus some- where in the Los Angeles metropolitan area. It would probably be more difficult for the Russians to find a bus anywhere in the freeway or roadway network in the Los Angeles basin than for them to locate one of these missiles in the shell game involving sophisticated military vehicles. But we have never addressed ourselves to that. The dilemma 78

of the Russian strategic planner is about the same as that faced by the Los Angeles commuter. The commuter cannot find a bus in the system, and does not know where or at what time it might pop up and be available to him. He is thoroughly flummoxed, and we hope the Russian planner is also. Thus we have gotten a very sophisticated $38-?0 billion invest- ment in one kind of system to keep our potential enemies guessing, yet we fail to address the problem of simpliflying a domestic transport system that seems equally confusing to our friends. That strikes me as one of the ironies of our current system. Another matter of concern in the institutional setting, referred to earlier, is the lack of predictability of many forms of research addressed in Garrison's paper that are to be carried out in the universities. Part of the reality at the present time is that the university is declining as a transportation research center. The uni- versities cannot count on very long-term support from UMTA, other parts of DOT, HUD, or other government departments for projects they may feel deserve a long-term exploration. Those in the Washington area have been through this process. At the invitation of DOT, five of the uni- versities in this area set up a transportation center. The DOT felt that since there were five universities involved, it must be five times as effective and therefore could operate with 20 percent of the funds that might be needed by any other single institution as long as we did not charge any overhead. So we wound up as vehicles to distribute the money to students so they could study with pro bono guidance from their faculty mentors on appropriate kinds of research. Many good products resulted, and the program had just begun to build up a suitable momen- tum—80 or 90 doctoral and master's level papers. Research projects had been carried through, and a number of young people had gone into the transportation industry. Then with about one-hour's warning, DOT told us that such a process would not be followed anymore, and that they were switching to "mission-oriented research." So the whole opera- tion simply stopped, as the basic source of funds switched to some other type of operation. That kind of disinvestment does not build confidence in research management or in the process of stimulation that might be provided. These processes need some rethinking and need to mature. Fortunately, it takes about four years for research management to mature both at the federal level and among the researchers them- selves. However, uncertainty tends to frustrate the researchers in both the public and the private sector. The time lag itself is another difficulty in the institutional setting for innovation. The last, and maybe the most fundamental, comment I would like to make in respect to Garrison's paper, and also to some of the others prepared for this meeting, is that there seems to be a lack of any real sense of the city as a complex with prospect and potential for an iden- tity that people seek. Instead, cities seem to be viewed simply as a cold-blooded aggregated market for indulging this, that, or the other economic analysis for adventure. We tend to see people viewing urban aggregation mainly in terms of market opportunities, but many cities have begun to reassert an identity. San Francisco, San Diego, Baltimore, Boston, and many others have broken the habit of simply regarding trans- 79

portation as the technological force to which they must adapt. They have grown to the point where they have begun to see the transportation systems as appropriate servants, within the limits for existing facili- ties they set for themselves, and for the purpose of aiding in the development of strategic locations. So part of the setting right now is a need to recognize that some cities have matured and would no longer build an elevated expressway down Atlantic Avenue, for example, as in Boston. No longer would they build an expressway along the waterfront as a path of least resistance. No longer would they necessarily locate truck terminals in places where relationships with residential areas would be disrupted. There is a certain maturity of expectations, not well articulated but undeniably present, in the minds of people who are getting sick of cities as places where they survive less and less well. We ought to think about that as providing not a niche, but a yawning gap within which much work needs to be done to think through how trans- portation systems relate one to the other and to the cities they are supposed to serve. But cities unfortunately are all too often seen, even by their nominal federal advocates, such as HUD or DOT or even Commerce, as more or less intractable problems. These problems add up to political liabilities. One does not want to get too close to these problems when elections are approaching. That is hardly the way to deal with the places where 80 percent of the population lives, and will continue to live, and which offer market opportunities and are more and more sophisticated and refined about where transportation and innova- tion might exist. It has to be recognized as a result of a limited and inadequate approach that the setting has not attracted a breadth of vision about these problems equal to the breadth of some of these ex- pectations that people have. Nor does the approach seem to be a reflec- tion of some of the values in addition to the opportunities for tech- nological experimentation that exist in urban places. That is part of the setting that is not addressed very much in most of the papers that I have seen. But again, I have discussed what is not there, as well as what is there, as a means of adding to the discussion. 80

INFRACTIONS OF GOVERNMENT, INDUSTRY, AND AfADRMTA REMARKS BY MARTIN GOLAND The topic assigned us is how to optimize the interactions between government, industry, and universities so as to take full advantage of these resources in advancing innovation in the broad transportation sec- tor. While the various transportation modes do have many technical features in common, each mode is also characterized by unique potentials and problems—unique by virtue of inherent function, historical develop- ment, and the nature of the market to be served. All of these factors must enter into our considerations. Our topic is thus a complicated one indeed. For the purpose of perspective, it may be of some interest to go back to an earlier and more traditional day and to recall briefly the relative roles played by universities, industry, and government in technological innovation. The universities were the fountainhead of research knowledge, producing the new scientific concepts and data that would later become the foundations for advanced industrial products and processes. Industry (and in this category I am including the individual in- ventor and entrepreneur whose ideas flowed directly into corporate in- dustry) represented the productive might of our nation: the producer of goods and services to meet market needs. Innovation within industry was a powerful force, because success in the marketplace depended on it, but it is fair to say that until the post-World War II period, industri- al innovation depended more on invention and on improved manufacturing technique than on science. Basic research was largely left as a uni- versity function, although there were a limited number of corporations who were the exceptions and who built strong scientific teams and research laboratories to meet their product improvement needs. The government role in those earlier and simpler days was a selec- tive one. In agriculture it is true that the government was the prime mover in a national program of unexcelled innovation and research ad- vance. Government laboratories also conducted research and development in other fields (and even some production in the case of government arsenals), but only in areas where industrial strength was lacking and where the government was essentially the sole customer (e.g., military equipment), or in areas where the nature of the activity was clearly a government responsibility (e.g., the Bureau of Standards). 81

If we compare the situation as it was some decades ago with that of today, it is clear that whatever order existed then has evolved into a far more complex system. Universities no longer monopolize the funda- mental research function. The government is actively supporting and conducting research and development in a host of nongovernmental market- place-oriented fields, using as its mandate a rapidly growing regulatory involvement and the emergence of energy as a critical national issue. Industry, while retaining its production primacy, has become a powerful force across the entire spectrum of technical activities ranging from basic research to societal analysis. The roles of the three performers have become far more overlapping and interrelated. This is one reason why this meeting can prove to be a valuable step toward achieving better understanding for future policy-making. One of the position papers is by Lawrence Goldmuntz. It is excel- lent, and it probes the question of what the government role in trans- portation innovation should be. In the past, there have been notable successes and failures. Our highway system is the finest in the world. The world fleet of commercial aircraft is still led by the products of U.S. manufacturers, and this preeminence would not have been achieved without the govern- ment-conducted program of the old National Advisory Committee for Aero- nautics, now the National Aeronautics and Space Administration. On the other hand, despite great financial support by the government, little innovative progress has been made in the field of urban mass transpor- tation, and government entry into the railroad field has produced little in the way of positive results thus far. Insofar as regulatory actions are concerned, the results are mixed—in some instances, the effects have been counterproductive because regulations were promulgated hastily and without the necessary foundations of knowledge and understanding of the problem. In this connection, incidentally, let me make the observation that in our assessment of the government role in transportation innovation, we should keep in mind that there are two "governments" in the picture, namely, the executive branch and the legislative branch. Whereas in past years the legislative branch has entered into technical affairs only weakly, this is no longer the case. Legislative committees have built up technical expertise as part of a major growth in staff capabili- ties, and the members of the House and Senate have individually become more deeply involved with technical issues. The legislative branch often takes the initiative in technical decisions, using its budgetary and law-making authority as levers. The executive branch, of course, has the responsibility for implementing legislative actions, and on occasion executive agencies find themselves the scapegoats for un- successful programs they did not devise. The larger government role has also affected both industry and university programs. Because of greatly increased regulatory pressures, industry has often been forced into a defensive posture, with meeting near-term goals a more urgent requirement than seeking longer-term objectives that may in the end be more worthwhile. The effects of such actions on innovation can only be surmised, but it is one area needing discussion. 82

Universities, for their part, have tended to become more applied in their research, not only because of a laudable desire to help in the solution of important societal problems, but also for the more pragmatic reason that money flows more easily in support of applied vs. fundamental research. Has this been a desirable trend in contributing to transportation innovation? No doubt our panels will have something to say along these lines. Perhaps in our panel discussions we will look for models from abroad. I think it is proper to say that in other technologically ad- vanced countries, universities are less directly coupled to industrial research than is the case in the United States. Research institutes, jointly supported by government and industry, are established to do research on generic problems that can be dealt with more cost-effective- ly by industry as a whole rather than by individual companies. The Japanese mode of operation involves an active government role well beyond that of merely financial support—along with industry, planning is done jointly in support of national initiatives and policies. While individual Japanese groups are intensely competitive with each other, their system nevertheless permits a degree of cooperation at the generic problem level that is not normally permitted under U.S. antitrust law. Finally, we will no doubt want to examine the levels of transporta- tion innovation in the U.S. compared with those abroad. Have other nations been more innovative than we? Railroads are more effective in passenger service in European countries and in Japan. Is this because of a rational response to a market differently constituted than ours, or is it because their approach is indeed more innovative? In automobile design, foreign manufacturers are frequently praised for being more innovative than those in the United States. Their cars are more energy-conserving, at least according to critics of the U.S. industry; they are better designed and are more serviceable. If one talks to automotive engineers abroad, however, their appraisal usually does not agree with these assertions. Many Japanese automotive engineers I have talked to admit that they have much greater experience in small- car design, certainly an asset in today's environment, by virtue of the fact that the Japanese marketplace has from the start demanded small cars. But they see their design advantage rapidly disapperaing as U.S. engineering teams concentrate their efforts in this area and as U.S. production facilities become better adjusted to the needs for small-car manufacture. Their continuing advantage, they contend, will be in the quality and dedication of their work force. The average Japanese pro- duction-line worker, in their opinion, is better educated, more interest- ed in the job, and instilled with the philosophy of insuring high quality in the end product. Those, combined with taxation policies that tend to encourage innovation and capital investment on the production line, are the strengths they will look to in the future. They do not foresee that continued Japanese success in automotive sales will be the result of superior Japanese innovation—some will even admit that U.S. pre- eminence will continue for some time. These, then, are some of the issues our panel might want to look into. I know that our discussions will be provocative and probing, and that we will give a worthwhile report. 83

INTERACTIONS OF GOVERNMENT, INDUSTRY, AND ACADEMIA BY DAVID S. POTTER PRESENTER: CRAIG MARKS INTRODUCTION It is generally agreed that innovation—be it technological, economic, social, or institutional—is one of the necessary ingredients in the pursuit of progress. However, for some years now, innovation has been discouraged in the United States, for reasons that are mostly indepen- dent of the technical field or industry. You have heard the litany of all management people, within govern- ment, industry, or academia, concerning the evils of inflation and high interest. Their application to technological matters is sufficiently specific, however, that I would like to address very briefly the par- ticular way that these economic factors affect technology and, through technology as one example, the opportunities for innovation. In a long-term sense, a good inflation strategy for an industry might well be a heavy early investment in future productivity as the means for transferring inflation-proof benefits into the future, and R§D expenditures offer such a possibility. In the short-term, though, inflation mitigates against such investments. Most manufacturers have found it impossible to recover inflation-caused cost increases totally. A 70 percent price recovery of inflation-induced cost increases is fairly normal. This shortfall results in a real shortage of capital funds for modernization or replacement of existing equipment. In addition to this first-order effect, there is a curious and trouble- some second-order effect so that there are even fewer R§D incentives than before. Successful new-product development would only lead to capital requirements for new products and processes that cannot be funded, so the "why bother?" attitude appears. Inflation-induced high interest rates also have an insidious effect on the willingness to accept entrepreneurial risks. With current high interest rates, the present value of any future benefit must be dis- counted so deeply that there is less incentive to undertake the higher- risk programs that characterize a dynamic economy. If the expectation value of a high-risk venture is less than the return on a government bond, prudent management does not invest in risk. Again, this situa- tion yields a second-order impact on R§D. Since one is looking either for very high payoff or relatively low risk programs, the number of R§D projects needed to explore the possibilities meeting these requirements is limited. If, however, the return on investment need be only half as much, as it was in the l940s and l950s as compared to today, then many more R5D efforts need to be undertaken because the cutoff point for 84

success has been significantly lowered. Even though high interest rates can be considered largely to be an internal U.S. matter, they ultimately have a profound effect on our ability to compete in international markets for years to come. A third area of general concern that is expressed most often by the industrial people is in the area of regulation. A certain amount of regulation is necessary, and I do not quarrel with the point of view that says that in our crowded society, central regulatory bodies are essential for the preservation of our environment and for other ameni- ties of life that we have come to require. No matter how good the regulation, however, it results in nonproductive expense, and that is inflationary. It is the thesis of this paper that a great deterrent to innovation in the transportation industry today is government inter- vention, much of it in the form of regulation. At the very least, the cost of the regulatory apparatus both within government and within industry must be factored into the price of the i product. I will return to this later. One also must comment that the uncertainties occasioned by regula- tion are at least as damaging as the regulation itself. The numbers change seemingly without cause; the test protocols change without notice; and the enforcement criteria change so as to, in effect, change the standard. Given these uncertainties, there can be a tendency to slow down and "wait and see." Let someone else be the guinea pig and learn the hard way what government meant by the latest change. The ultimate problem with innovation in the private sector is that it exists within, and responds to, all of the same stimuli as its sponsor, the business community. The problems of the private sector eventually all depend upon the health and well-being of our nation's economy. No matter how well intentioned government is in the encourage- ment of innovation through the many mechanisms available in the federal system, those efforts can have little real impact on the citizenry within our presently conceived private enterprise system. At the risk of too much repetition, I want to underscore the fact that the applica- tion of science and technology to the well-being of the people is a function of the private sector. Ultimately, the only way to keep private science and technology healthy is to heal the economy. Therefore the climate for innovation in the transportation industry requires the essential ingredients of favorable economics, an identified need and good ideas. It requires an inventive and aggressive state of mind on the part of the various participants be they business people, inventors, purchasers and users, or government policy-makers. And, importantly, in the advanced and mature state of our society, innova- tion requires a mutually supportive attitude between and by the parti- cipants in the process. The balance of this paper, then, is directed to developing these points in the context of transportation. The first section discusses the opportunities for innovation in transportation as a function of the stages of development—the new, developing, and mature stages. The second section discusses the roles that government has played in the encouragement or discouragement of transportation innovation— 85

first, as a sponsor, purchaser, and user of the results; second, in the development of specifications for transportation products or processes for use by others; and, finally, as a regulator. With more and more departure from free market behavior and with increasing levels of interference by government, innovation wanes and disappears. Inappro- priate government becomes the major constraint to innovation in trans- portation. THE STAGES OF DEVELOPMENT IN TRANSPORTATION The automobile industry can serve as a useful example of the potential differences that exist in opportunities for innovation as a function of the stages of industrial growth and development. In the new or initial stage, there is the introduction of a new product or service. There are many participants. Entrepreneurs and inventors are prevalent. There is rapid growth in application or sales. There is high opportunity and, particularly for the individual entrepre- neur, risk. In the past, government has had little or no role. Basic knowledge growing out of academic studies has potentially high applica- tion and utility. The automobile industry had its beginnings in the late l800s and early l900s. The automobile did not spring, full-blown, into existence but evolved from a variety of dreams and ideas for self-propelled vehicles. The development of practical, liquid-fueled engines around l885 gave impetus to vehicle development so that by 1900 in the United States there were some 8,000 crude automobiles operating over dirt roads and brick and cobblestone city streets with more than l50 active automobile companies. The number of companies grew to over 200 by l903, with 22,000 vehicles produced in that year. Seventy-seven thousand vehicles were registered in l905, and almost half a million in l9l0. The people involved at this stage came from an interesting variety of backgrounds — inventors, engineers, blacksmiths, carriage and wagon makers, bankers, businessmen, and motoring enthusiasts. Laissez-faire conditions prevailed. Innovation was widespread. For example, propul- sion choices included battery electrics, steam, diesel, and gasoline engines. Steering could be by tiller or by steering wheels, gears, and linkages. Risk was justified or rationalized by the entrepreneur because of the high potential return. Newspapers, magazines, and trade and technical journals devoted exclusively to the automobile appeared and multiplied. A few speed limits and traffic regulations specific to the automobile began to appear. In the earlier developmental stages, the product or service of an industry is improved, expanded, extended, and perfected. Improvements occur both in the development of the original technology or idea and in the replacement of the original technology with a better technology. Standardization commences. Economies of scale appear. Competition 86

becomes even more intense. Economic risks become even larger. Govern- ment interest grows, and initial government controls may be applied. Needed regulation often was in the form of self-regulation. Thus standardization commenced on a voluntary basis. The Society of Auto- mobile Engineers (which was first formed in l905 and became the Society of Automotive Engineers in l9l7) created a Standardization Committee in l9l0 and succeeded in reducing the number, kinds, and sizes of the various parts that went into an assembled automobile. Gradually, and without government intervention, steering wheels were standardized on the left side; gear shift patterns and gas pedal, brake, and clutch positions were standardized; and such items as four-wheel hydraulic brakes and windshield wipers became standard equipment. The dominance of gasoline-engine-equipped, closed-body automobiles was established. The Federal Aid Road Act, in l9l6, was the beginning of a national system of interstate highways. Wartime excise taxes were levied on automobiles in l9l8 as dispensible luxuries. Automotive exicse taxes and fuel taxes were introduced, extended, and increased over time. Opportunities for innovation began to shift toward process and productivity improvements (to the advantage of the consumer) although substantial product innovations continued to be introduced. Several examples include four-wheel brakes in l924, safety glass in l926, and automatic transmissions and turn signals in l939. It should be noted that the evolutionary improvement of the auto- mobile was directly recognized and applauded by the continued enthusias- tic response of the consumer. Sales and use of the automobile continued to increase on a long-term basis. As development continues, the product or service becomes establish- ed. Customer or user expectations are extremely high. Product advance- ments must be complemented by innovations in other areas, e.g., decreas- ed overhead or manufacturing costs. Competition is very keen. Oppor- tunity remains high, but risks are also very high. Product or process standardization may occur. However, government regulation can become a dominant factor in both business operation and product design. This can result in a narrowing of the scope for possible innovation and can encourage focusing more on reducing risks than on venturesome excursions into the unknown. Automobile industry developments during the period from World War II to the present represent a period of consolidation for some companies and expansion and modernization for others. Despite the tremendous regulatory burden, competitive forces continue to result in innovations. A few examples are plastic dies for steel stampings in l952, widespread availability of power steering, also in l952, power brakes in l953, im- proved sealed beam headlamps in l954, standardization of amber lights for front turn signals in l962, self-adjusting brakes in l963, and car warranties over several years and the energy-absorbing steering column in l966. At the same time, government interaction with the industry also developed. The Interstate Highway System was approved in the Highway Act of l956. The Highway Trust Fund was established to apply highway user taxes to finance the federal share of the programs. In l966, the 87

National Traffic and Motor Vehicle Safety Act and the Highway Safety Act were passed, and the U.S. Department of Transportation was formed. Motor vehicle safety standards were established and enforced. The Clean Air Act was passed in l970 and amended in l977. Standards for automo- tive exhaust emissions have been established. The Energy Policy and Conservation Act was passed in December l976, and it mandated automobile fuel economy standards for l978 to l980 and for l985. Other Examples of Opportunities for Transportation Innovation as Functions of the Stages of Development A brief mention of examples of opportunities for innovation in other segments of the transportation industry is in order. Rail Locomotives As railroads have developed in the United States, the cost of labor to operate the trains has been historically high. The introduction of the air brake, improved rail car couplers, improved brake shoes, and impor- tantly, the vast improvement in control of starting tractive effort made possible with the diesel-electric locomotive all permitted rail- roads to increase the length of freight trains and thereby to minimize operating costs. Through these and other advantages, the diesel-elec- tric locomotive has become the dominant form of motive power for rail- roads in the United States. Now, with energy (both cost and availability) considerations becom- ing of growing importance, rail companies are taking another hard look at electrification and alternative energy sources for powering trains. But given the magnitude of the needed investment (especially in the light of the financial plight of many of the railroads), competi- tion for available funds by other needed rail projects, an absence of needed information on potential service or reliability advantages of electrification, and environmental questions—not to mention rate of return on investment considerations—make electrification an extremely high risk issue. The Northwest Corridor Amtrak facility is presently scheduled for improvement of the existing electrification between Washington, D.C., and New Rochelle, New York, and new electrification is slated between New Haven, Connecticut, and Boston, Massachusetts. No other major railroads have anything specific in hand in the way of studies or plans. Transit Buses During the history of the motor coach business, a number of major im- provements have been introduced and made standard by the industry. Included are diesel-engine-powered motor coaches, air suspension, air conditioning, automatic transmissions, integral aluminum bodies, 88

transverse rear engine mounting, and fluorescent lighting. (We are happy to note that GM led in pioneering these developments.) In the early l970s, in response to transit authority demands for new bus equipment, General Motors announced its plans to produce the first all-new bus design since the "new look" bus was introduced by General Motors in l959. The new General Motors design, called the Rapid Transit Series (RTS), not only reflected an entirely new modern-day look and design, it was to be built by a completely different advanced manufacturing process. The U.S. Department of Transportation was notified of the availability of the RTS in early l97l. However, DOT refused to alter its procurement policies to permit purchases of this advanced design bus with federal assistance. Instead, the department asked bus manu- facturers to participate in a DOT-sponsored effort, announced in mid- l97l, to develop a new government-sponsored bus design under the project name "Transbus." What has happened since is history. The government's Transbus program was conducted, at great expense and at the cost of delaying by several years the introduction of the new design GM had ready to go. There have been changes in transit bus procurement procedures and both Transbus and RTS specifications. There have been changes in administra- tions and U.S. Department of Transportation personnel. In addition, litigation further delayed introduction of the new bus design. Deliveries of the GM's new-design coaches finally were allowed to begin in September l977 but not without continuing difficulties in ob- taining satisfactory government procurement procedures. The issue of whether the government will attempt to issue a mandatory bus design to be followed by all producers is just now being resolved. Needless to say, these events will have considerable impact on the opportunities for innovation in transit bus design and manufacture. Furthermore, the availability of new equipment, deemed by transit operators to be needed, has been delayed by six years. THE ROLES OF GOVERNMENT That brings us to a discussion of the roles that government has played in the encouragement and discouragement of transportation innovation and to the development of suggestions for improvements in the govern- ment role. Government has several, sometimes conflicting, responsibilities in transportation. On occasion, government agencies are customers for equipment or services. Sometimes they are the developers of specifi- cations for equipment to be purchased and used by others—public agencies agencies and private individuals. Finally, government has the responsi- bility to protect the public interest in a very broad social context. In those several responsibilities, government has a key role, to provide a supportive climate for the encouragement of innovation in transportation. One major deterrent to innovation in transportation is the adversary relationships that have developed between government as 89

the regulator, the transportation industry (carriers and manufacturers) as the regulated, and shippers as the affected customer. These adversary relationships have been escalated by the prevail- ing public perceptions of big government and big business. As a result, regulation has become unilateral, with no room for experimentation or innovation. Some examples are in order. Government as Customer One example of government as an encourager and purchaser of innovation has been in the Department of Defense (DOD). Since another panel of the workshop is addressing procurement incentives directly, only a few points will be made here. There are many examples that might be cited, but we will take the example of the Main Battle Tank, in which General Motors played a role. This program, begun in the mid-l960s was for a completely new piece of military equipment with operational characteris- tics that far exceeded anything available at that time--a real step forward in the state-of-the-art was required. The original Request for Proposals (RFP) from DOD included performance specifications and was written in such a way as to invite innovation. The RFP spurred an active and healthy competition among potential contractors. During the research and development contract preparation period, there were numer- ous interactions and improvements of the subsequent contract performance specifications. Importantly, the contracts were written in a way that specifically invited and rewarded the contribution of innovative ideas and products at all stages in the overall design and development process. Government as a Developer of Specifications for Others As an example of where government has not performed quite as well, we will continue the transit bus story begun earlier. The new GM transit bus designs completed in 197l were designed to be superior public transit equipment, attractive in appearance and economical to operate and maintain. They were designed to respond to the functional demands of the transit marketplace, with many new, ad- vanced design features. In investing in tooling, GM expected that the transit system operators—whose rider's needs were considered in designing the vehicle- would be allowed to consider cost-effective features in determining the lowest evaluated bid. Only by allowing credit for the increased values of the new design could it be on a competitive basis with existing, un- improved models. DOT refused to allow full credit for such cost benefits, and the ability to market the advanced design buses has been impaired. Indeed, much of the time, DOT has insisted on contract awards to the lowest bidder regardless of quality. 90

Requiring bus contracts to be awarded on the basis of initial cost alone is at odds with fundamental, well-established procurement policy. Imposition of such a requirement discourages innovation and tends to push awards in the direction of the apparently "cheapest" (lowest ini- tial cost) equipment. The inadequacies of a low-initial-cost policy have long been recog- nized by the federal government in its own procurement regulations. There regulations declare that "the award of a contract to an offerer solely on the basis of the lowest evaluated prices is a disservice to the government if subsequently the contract defaults, is late in deli- veries or otherwise performs unsatisfactorily." The same regulations further provide: "While it is important that purchases be made on the basis of offers which are most advantageous to the government, price and other factors considered, this does not require an award to an offerer solely because he submits the lowest bid or offer." The Armed Service Procurement Regulations also expressly endorse the best-value, life cycle cost concepts. The government's Office of Management and Budget has taken steps to assure that federal grant recipients make their own procurement decisions. This general policy is expressed in the Uniform Administra- tive Requirements for Grants-in-Aid to State and Local Governments (OMB Circular A-l02) that stresses reliance on local initiative and declares in favor of a policy of "greater reliance on state and local governments." Despite this statement of federal policy, which has been in effect since l972, DOT for the most part has not allowed self-determination by local agencies and has preempted transit operators' procurement policies and practices. The low-initial-bid requirement finally was replaced by DOT with a policy permitting contract awards to the supplier submitting the lowest evaluated bid, with local transit operators making the comparative evaluations of competing vehicles. This step, taken in l976, was in- tended to give state and local grantees the same right to consider life cycle cost benefits as the federal government exercises in making its own purchases. However, before this new bid procedure could be implemented, the then secretary of transportation, shortly after taking office, issued yet another procurement policy. Under the current procedure, awards are made to the supplier submitting the lowest adjusted bid. Price off- set credits are allowed for some (but not all) superior, cost-beneficial features in determining the lowest adjusted price, but with a key difference from the l976 policy. Under the current system, DOT offici- als, not the transit authorities, make the evaluations. Under this policy, DOT'officials have disallowed price offset credits for features that transit authorities value highly and for which they have proposed to allow credits. In another of its many reversals of policy, DOT in l977 directed that initial purchases of the Transbus design that DOT mandated for all manufacturers be made on the low-initial-bid-price basis, returning to 9l

the philosophy that makes no allowances for real value. Subsequently, after extensive review, DOT, confronted with the many design and opera- tional difficulties inherent in the Transbus design, decided to post- pone Transbus indefinitely. The findings of a review panel selected by the government completely support the bus manufacturers' decision not to build the Transbus design as sound business judgment based on the problems found in the design. Bus manufacturers and transit authorities alike hope that the recent endorsement by Congress of the life cycle cost concept in the Surface Transportation Assistance Act of l978 will be fully implemented by DOT. The related provision states that after September 30, l979, transit rolling stock procurements can be based on consideration of life cycle costs and factors other than initial cost. Manufacturers cannot continue to spend time and money on projects that change every two or three years and for which there is no opportu- nity to recover costs. In l968 a National Academy of Engineering report recommended that the role of the federal government in transit equipment purchases should be one primarily of establishing technical criteria. However, as the l978 Office of Technology Assessment (OTA) report ("An Analysis of Urban Transit Vehicle Development and Demonstration Programs") pointed out, this procedure was never adopted. The OTA report found that DOT has injected itself deeply into the bus design business, almost to the exclusion of the manufacturers and the local transit authorities. The result has been a system that tends to stifle competition and fails to stimulate the equipment development that comes from a free market rela- tionship between manufacturer and customer. One possible alternative for dealing with this situation is a mechanism by which the users of transit equipment can play a greater role in determining specifications, cost effectiveness and performance of transit equipment. DOT's role should be that of a monitoring, con- sultative, and administrative agency. Government as a Participant in Financial Support of Research and Development as Potential Contributors to Innovation Members of industry usually are reluctant to posit a governmental role in the "free" economy. Let me stipulate at the outset that government has a right and an obligation to examine those facets of science and technology that have a direct impact on the health and well-being of American citizens and institution? and to provide funding where appro- priate. Historically, most support for scientific activities in this country has come from the intellectual interests and philanthropic activities of private citizens. Technological development largely was carried out by the private sector in response to ordinary commercial incentives to create a new or better product or to create an equivalent product at a lower cost. The direct oversight of government in these matters was not so necessary as it seems today, and except for some 92

early military development, there was little federal funding. Congress did recognize national interests and exhibited great leadership at a time of real need in developing such programs as the national land grant college program in l862 through the Morrill Act. As evidence of this program's ongoing contribution to science, one can cite the former American Council on Education evaluations of academic excellence. In addition to the program's intended impact on agriculture and mechanical studies, land grant institutions include some of the nation's finest scientific and technical organizations. Congress certainly played a leadership role and accomplished a great deal over the years with this fine program. Soon thereafter, in l863, Senator Wilson of Massachusetts drafted the bill that has become the charter of the National Academy of Sciences. To the everlasting glory of Congress, it has seen fit to leave it in its original form, unchanged, to this day. After World War II, it was necessary to continue federal support and recognition of science. In describing the creation of the National Science Foundation, Vannevar Bush said, "To persuade the Congress of these pragmatically inclined United States to establish a strong organi- zation to support fundamental research would seem to be one of the minor miracles." This minor miracle did occur and in the postwar era the Congress not only accepted the idea of a National Science Founda- tion and was instrumental in getting it established, but also pressed for the large growth of the National Institutes of Health. Other exam- ples of creative congressional action include the Atomic Energy Commis- sion, the Sea Grant college program, and the National Aeronautics and Space Administration. Opinions differ on the continuing necessity and effectiveness of these activities, but nonetheless, they provided solu- tions to perceived national problems. In other areas, the Congress has assumed a more passive role in research and development activities within existing departments and agencies of the federal government. Let me now address the R§D activities that are directly funded through the federal government and hence offer a direct opportunity for congressional involvement and policymaking. An important congressional concern should be, and has been, main- taining a healthy scientific establishment in this country. This means maintaining institutions and educational facilities to produce research personnel of sufficient quality and quantity to serve the needs of the nation. In the process, of course, scientific research is generated. In discussing the training necessary for science and technology and also the research results that are generated in the education process, it is difficult to sort out whether trained personnel or research results are the primary product. The National Science Foundation and the National Institutes of Health seem to have achieved a reasonable balance between the two. The professionals on the scientific side would certainly argue that Congress should have been more generous in its support and one would also guess that the specific allocation of money to the various grantees must be thought to be somewhat less than perfect by those who failed to 93

receive grants, but such complaints are certainly to be expected and do not detract from my general observation of a well-considered and executed program by these institutions. It also seems, in retrospect, that congressional handling of the budgetary process has given some sense of national need and priorities to these institutions. Although Congress tends to become quite specific in management objectives in some of the larger programs, for instance the Mohole project, or the work in Antarctica, there seems to have been a general appreciation that Congress in its role of "Board of Directors" is better off giving policy guidance, establishing priorities, and insisting on good manage- ment than attempting a detailed management of the enterprise. The excursion of the National Science Foundation into the more applied world via the RANN (Research Applied to National Needs) program was to my mind a mistake. That excursion, however, was one suggested by a past administration and not by Congress, and I am glad to see that the experiment has ended. As one moves from the support of science for its own sake to the support of science because of the needs of some mission-oriented agency of department, the record becomes more spotty. In addition, the fund- ing procedures become more complex and variable. In a meeting of the American Association for the Advancement of Science last year on R§D in the federal budget, John C. Calhoun of Texas A§M observed that the R§D budget appears to consist of three processes: one at the agency level, one at the executive policy level, and one at the congressional appro- priation level. Each appears to have its own ground rules. Although the processes overlap and are intertwined, they appear to be based on separate analyses and different assumptions. My personal experience with the Department of Defense and the Office of Naval Research only served to reinforce that view. I am a firm believer in competition and have generally approved of the fact that a researcher might well have two or three sources for funding. In dealing with a bureaucracy, which I will define as a group of profes- sional managers who in trying to achieve some overall good result will not personally be affected by a bad result, it is essential that multi- ple paths be provided. Although I am no longer personally engaged as either a donor or a beneficiary in government research grants or con- tracts, I am disturbed by an increasing whimsy on the part of some government agencies in making grants. It seems to me that a Mansfield- type amendment, which would restrict the kinds of programs that can be funded by mission-oriented agencies, promotes short-term efficiency at the expense of a long-term loss of vitality. Across the full spectrum of the science and technology that Congress supports through direct funding programs, I would conclude that, in general, the science program of this country has been shaped by the national priorities and instructions given by Congress through their allocation of resources to various sectors of science and technology. The congressional choice of the priorities and the consequent shape of our scientific community may be judged successful by some and a poor compromise by others, but the accomplishments and the imposition of the "sense of the Congress" on the scientific establishment should not be doubted. 94

Now I would like to direct the discussion into those large areas of science and technology that do not receive direct government fund- ing; those that are carried out and financed by the private sector. In this area, Congress has a responsibility to create a climate that nur- tures research and development efforts within the private sector and to monitor the nation's health in this area from time to time. It is only through the private sector that goods and services are provided to our people. Government-funded R§D, in and of itself, can- not provide a continued high standard of living of our people and ade- quate employment opportunities for all. In recent years, we have wit- nessed a relative decline in the American economy. Although in an absolute sense we have retained a strong economy, we have lost ground to other important economies in the world. But more importantly, our rate of improvement has not kept pace with our own expectations. We have the scientific, technical, and managerial capacities to do better than we have. That brings us, then, to a summary discussion of actions that might be considered by government, generally, and by the U.S. Department of Transportation, specifically, to regenerate the needed supportive cli- mate for innovation in transportation. Above all else, government must take the actions necessary to return the national economy to a healthy state—to eliminate inflation and lower interest rates. Economist Milton Friedman stated the case as succinctly as possible when, in commenting on the growing public dissatisfaction with government, he said, "...inflation is produced primarily in Washington." It makes very little difference what government does to improve the climate for innovation until the national economy improves. But given a healthy economy, there are some specific things that can be done for the transportation industry. Cooperative Automotive Research Program One example is the Cooperative Automotive Research Program (CARP) presently being negotiated between government and industry. On May l8, l979, representatives from the domestic auto industry met with President Carter to discuss a basic research initiative. While the industry agreed to the principles embodied in the intiative, the details remain to be worked out. The initiative closely parallels the program presented by Philip Smith, Associate Director, Office of Science and Technology Policy, in testimony on May 2, with respect to H.R. 4678, "Automotive Research and Technology Development." The seven stated objectives of this legislation are as follows: To preserve and enhance personal mobility at reasonable cost To reduce dependence on imported oil To increase motor vehicle safety To reduce motor vehicle environmental effects To improve motor vehicle reliability ° To conserve scarce resources 0 To enhance the international competitive position of U.S. autos 95

If we were asked to list the major objectives of our own programs independently, it would be difficult to improve on those listed in this legislation. However, government involvement in automotive research should be focused on basic research, not component or product development. This is a very important distinction that is crucial to a successful industry-government relationship. Admittedly, the bounds of basic research are not always clear, but basic research can be roughly defined as an effort to discover new knowledge without specific commercial objectives. Technology develop- ment, as defined in Section 502 of the legislation, goes well beyond the discovery of new knowledge and would have the government working on a parallel path with industry by authorizing the government to "devise new component and system concepts, and develop new experimental compon- ents, subsystems, and vehicles when necessary to verify such concepts." This authority is likely to encourage government to enter into areas of product development that would duplicate industry efforts and be wasteful of national resources. The auto industry has demonstrated the capability to produce the hardware necessary to provide economical, safe, socially acceptable transportation within the framework of manu- facturing, marketing, and financial constraints. I believe that tech- nology development—the application of the results of basic research-- should be left in the hands of industry. Basic research, on the other hand, is a legitimate area for govern- ment involvement, and there is a wide range of automotive subjects that would seem to qualify for basic research under government auspices. The previously discussed initiative prepared by the Office of Science and Technology Policy proposed the following subjects: o thermodynamics, combustion, and fluid dynamics o structures o noise and vibration 0 materials science and processing o control systems o friction and wear While there may well be others, each of these are important areas that would benefit by an expanded research effort. Basic research, which has no specific commercial objective, typically benefits society in general as well as those industries using the resultant fundamental knowledge to improve their products. In the case of the auto industry, the incentives of the marketplace are more than sufficient to encourage application of knowledge from any source to improve our products. It is recommended therefore that the necessary funding for basic research programs be obtained from general revenues, as is the case with government research in other areas. It is recognized that an effective basic research program, along the lines indicated here, will require an organized structure within which to implement and coordinate the work to be undertaken. It is essential that, whatever existing government agency is selected to undertake an independent basic research program, high standards of leadership be established. The head of the organization should be a 96

person appointed by the president, subject, of course, to confirmation proceedings. The criteria for appointment should include stature in the scientific community, a record of achievement, and demonstrated com- petence in the disciplines of basic research. In addition, staffing for the program should include individuals with a demonstrated ability to manage basic research projects effectively. It is also important that an independent advisory board be includ- ed in the organization to assist in the determination of programs to be undertaken. Further, provisions should be made for third-party peer review of research procedures to be certain that parochial interests are minimized and that objectivity is maintained in all respects. The Department of Transportation should be considered for at least a portion of this basic research program, provided the organization is structured and staffed along the lines described above. This would mean that research would be under the direction of a competent, highly qualified scientist at the assistant secretary level. This would not only serve to upgrade the needed scientific capability within DOT, but also help to assure that any research project that is undertaken would meet rigorous scientific standards and would be in concert with national needs and priorities. Finally, I have some general thoughts on government support of basic science, research, and development. On an overall basis, I believe that the Congress deserves high marks for much of its perfor- mance in its role as overseer of U.S. science and technology. On many occasions, Congress has anticipated the scientific community's needs in our country and has provided farsighted leadership in meeting those needs. The best performances seem to have been in those areas of direct federal funding for science in the abstract and in those situations in which science and technology are funded to support an important field wherein the practitioners are highly fragmented, as in agriculture, fishing, and health care. As one proceeds from the support of science to the more program- matic endeavors, the record becomes less impressive. Congress seems to operate best when it undertakes the role of policymaker. When govern- ment assigns priorities by major budget allocations into various fields of science, when it insists upon good management by the agencies involv- ed, and when it stays out of the detailed management of the business, the enterprise prospers. In this way, our R§D budgets do tend to track the larger needs of the country as perceived by the Congress. This is as it should be. However, government is ignoring a pressing need for good basic science. Scientific research for regulation, especially health-related regulations such as the Clean Air Act, is woefully inadequate. The Clean Air Act was written in the middle l960s and was based on health- needs data that were sketchy at best. And yet, more than a decade later, the body of air-quality-related health data is virtually unchanged. Congress has a golden opportunity to serve the public by supporting such work from institutions that are outside the regulatory structure. The research programs designed to resolve the controversy surround- ing the issue of diesel particulates also illustrate the degree to 97

which regulatory bodies and industry could coordinate their separate operations toward insuring that scientific efforts pursue the true public interest. In the past, regulatory decisions have sometimes been based on incomplete data. The nature of judgment is such that in some situations, reasonable individuals honestly disagree with the ultimate decisions. For example, there is an urgent need for data on the nature and possible impact of diesel particulate emissions well before the regulations are promulgated. Coincident with that is the obvious desire of all involved in the eventual decision-making process to have the meaningful and appropriate data that are necessary for the deci- sion in their own hands. To resolve this type of dilemma, General Motors proposed a simple and yet unique approach. Instead of having each party design and conduct its research program and then wait until the actual rule-making procedure began to tell the other what the results of its research programs had been, GM and EPA have had a series of discussions in which they exchanged their plans for the diesel par- ticulate research they intended to conduct. While neither organization is in any way constrained from duplicating the other's efforts, it is hoped that any planned duplication would be both conscious and purpose- ful. The ultimate goal of this early exchange of plans is to assure that when the rule-making process reaches the point of decision making, relevant research will have been done by one or another of the partici- pants. This can only result in regulations that are more intellectually sound, that will be less controversial, and that will best serve the overall public interest. In addition, the National Academy of Sciences and National Academy of Engineering were asked to establish a panel to review the diesel particulate situation in the light of current regulations and available knowledge. Their charge is to make recommendations on courses of action. This independent, third-party review should prove to be constructive. Government as Regulator As was stated at the outset, no rational individual would seriously argue with the idea that in our crowded society central regulatory bodies are essential for the preservation of our environment and for the other amenities of life that we have come to require. No matter how good the regulation, however, it results in nonproductive expense, which is inflationary. Although somewhere, somehow, it is assumed that a benefit will be achieved for some segment of our population, the effect of a Clean Air Act, or any other regulation of this type, is to increase the specific cost to the customer. Since no extra benefit usually is conveyed to the one individual consumer for that particular unit or service, the cost increase is identical with inflation. Somebody must pay for the l0,000 federal employees of the Environmental Protection Agency and for the tens of thousands of people in industry and in government who must correspond and interact with those l0,000. The greater impact, in the context of innovation, is when the commensurate benefit to balance 98

the cost is not received by anybody. Those regulations that achieve marginal benefits of low value do more than waste money. They also divert resources that could otherwise be available for the innovative entrepreneurial activities that the nation needs. The diversion of human resources is perhaps the greatest cost of all. A more direct impact on innovation from regulation stems from the complexity and scope of the current regulatory system and the extreme severity of the penalties for noncompliance. The ultrahigh risks to a manufacturer of noncompliance—risks that range from jail terms for executives to huge fines for corporations--tend to stifle innovation by reducing both the range of potential innovations that are extensively developed and the rate at which innovations can seriously be considered for implementation. A potential innovation that promised a potentially high "pay off" in terms of customer acceptance and a reasonably good chance for successful performance in the field would stand a good chance for implementation in an industry where market forces were the sole disciplinarian. However, in a highly regulated industry where severe penalties would be added to the penalties of the marketplace in the event of the innovation's failure, chances are that prudent manage- ment would not implement the innovation without a great deal of addi- tional development. Under the best of circumstances, this situation means that the rate of innovation is retarded by the regulatory system. In practice, such a highly punitive system virtually guarantees that product development is more evolutionary than revolutionary. The ulti- mate result of regulation can easily become a standardization of pro- ducts and services, rather than a wider variety of consumer options. One branch of government that sometimes is not generally regarded as a regulator is the judiciary. Over the years, though, the courts have become a significant consideration in the design process of most industries, including transportation, and also in the decision-making process for the introduction of innovative product changes. Under the current conditions, it is possible that in some courts a design improve- ment in an existing product could be construed as an admission that previous designs were less than optimal and could subject the manufac- turer to product liability claims for not having included the newer design on earlier models. Clearly, this is not a climate that is con- ducive to innovation. The least onerous result of an overly broad con- cept of product liability is that products cannot be designed to be used by the normally prudent and thinking consumer. Instead, they must be designed to protect the reckless and thoughtless. This usually means a penalty—either in cost or in utility—for the typical customer. (One should note that this mentality is in no way restricted solely to the judiciary. The case can be made that the decision to require passive restraints in automobiles pivots upon the same philosophical point and penalizes the prudent auto occupant who already is utilizing existing restraint systems.) So far, we have largely dealt with the effects of government actions on innovation within existing industries and technologies. It should be noted that regulations, which usually are enacted to deal with per- ceived problems within the existing scheme of things, can virtually preclude innovation from directions that were not foreseen at the time 99

the regulations were conceived or for needs that emerge after regula- tions are in place. Regulation also can impede the innovative process by requiring that fledgling innovative technology compete on the same basis as mature technology. In the absence of regulation, the innovator could identify a niche in which the dominant technology was not the optimal approach and innovate to compete within that niche. If the innovation were successful, it had the potential to grow, improve, and expand into other economic niches in a Darwinian fashion. Regulation usually applies universal criteria over such broad areas that many niches no longer exist. While the need for an innovation still is there, broad regulation can preclude innovation for limited application. Obviously, limiting the potential application for innovation also limits the im- petus and the likelihood of innovation. Congress recognized this problem during the creation of the Clean Air Act and its potential impact on the development of diesel engines. It also recognized that the diesel engine, because of its potential contribution toward energy conservation, might require some temporary exemptions from certain provisions of the Clean Air Act—specifically from the NOX emission requirements—and included provisions for granting a NOX waiver for diesels. As it turns out, such a waiver probably will be necessary if General Motors is to be able to offer diesel engines beyond the l980 model year. Although such a waiver is not automatic— it requires the demonstration of good faith efforts and the absence of danger to the public health—it represents a regulatory feature that is missing in most regulations. It allows time for a young innovation, light-duty, clean diesels, to develop into a competitor against a well- developed technology with a long history of evolution, the light-duty gasoline engine. Assuming that the waiver is granted (as of this writing, the waiver has not been granted), Congress will have success- fully avoided two common pitfalls of regulation—the inadvertent ex- clusion of innovation from unexpected directions and also the accomoda- tion of conflicting regulatory goals, namely emission control and fuel economy. But even if things do go smoothly with the diesel, the situa- tion represents a good example of the potential restrictions of the regulatory process and illustrates how the process easily could prevent-- or at least discourage — innovation. There are other examples where the negative effects of regulations were not foreseen and moderated. One potential energy source, shale oil, is becoming increasingly attractive as petroleum prices rise and as the U.S. balance of payments problems are exacerbated by petroleum imports. Yet, it is virtually impossible for a private enterprise to undertake the task of building a pilot plant to develop the technology to produce oil from shale. It is impossible because of environmental regulatory restraints. Under today's regulations, such a pilot plant would have to guarantee that no matter what technical problems were en- countered and no matter what problems might occur, the plant would comply with all environmental regulations. Clearly, when one is deal- ing with the task of literally inventing a technology, compliance can- not be guaranteed with l00 percent certainty. But given the ultrarisks _j l00

of noncompliance, a prudent individual or corporation could not reason- ably undertake to develop such a technology even with the honest and unreserved intention of making the best possible effort to maintain and protect the environment. Furthermore, if an individual or organization were willing to take such a risk, the regulatory/judicial system simply would not allow ground to be broken without the demonstrated ability to meet all contingencies. In other area, our political system is equipped to handle situations where there is the likelihood that a law is about to be broken. In most cases, under the provisions of the Constitution and/or the Bill of Rights, the government's power for such "prior restraint" is strictly circumscribed. Such limits on the power of government emerged after decades of political, moral, and ethical discussions that drew on the intellectual and political resources and experiences of academicians through the ages. Our current regulatory system includes some of the fine features of our political system—due process, the rights of individuals to petition and to be heard—and any reasonable person would argue that those rights ought not to be denied to anyone. But at the same time, it is apparent to the reasonable individual that much of our regulatory system evolved on an ad_ hoc basis and that in its totality, the system lacks a cohesive and consistent philosophical and ethical framework. Perhaps the real need is to generate a climate where the academicians from all disciplines and the bureaucrats on all levels can honestly raise philosophical and ethical questions about the regulatory system without being criticized for abandoning their respective charters and develop a system that would allow some of us to teach and all of us to learn how to regulate better. If this were to happen, it would be the best possible innovation of all. DISCUSSANT'S COMMENTS BY JOHN G. TRUXAL The previous paper summarizes concisely and effectively the basic problem in the development of the role of academic institutions in inno- vation in transportation. The paper focuses primarily on the current "state" of this system and the historical background that has led us into the situation in which universities (and nonprofit research institutions) are playing a relatively minor role in the transportation field, in which the conflict between industry and government (especially in the regulatory arena) is aggravated by a deep-seated public distrust of both government and industry. From these arguments, however, the paper does not offer any con- structive approaches on how this situation might be changed. In my brief comments, I would like to try to offer two specific suggestions in the hope that the panel deliberations will emphasize such a look into the future, rather than a lamentation on the past. l0l

First, in the area of research, the paper calls for a strengthen- ing of basic research (and a deemphasis of applied research). I am not sure the distinction is so easy to draw, but my principal reservation is that the author overrates the quality and significance of present university research programs in transportation. I believe this current activity is not a strong element of the total national picture in en- gineering and science research. There are several reasons for this: l. The government (i.e., DOT) has not evolved a consistent, steady policy encouraging participation by leading research personnel. 2. There is no long-term, continuing effort, similar to the Joint Services Electronics Program of the Department of Defense—no program that promises stable support as a university group develops competence and experience. 3. There are too many short-term small grants (which tend to attract individuals who are not supported otherwise). While some of the researchers are excellent, others are not, and the transportation industry does not attract its proper share of the gifted young people. Second, the author makes a strong issue of the public distrust of both government and industry in transportation. He focuses superbly on this situation as the core of the problem, but I would have added that correction of this attitude is clearly an appropriate use of academic institutions. In the regulatory area, for example, there is practically no public education—no careful, in-depth analysis of the goals of a particular program. As one example, the corporate fleet mileage goal for the l980s, discussed so widely in the mass media, are mysterious to the educated adult. Are they significant in terms of petroleum consumption? Are they placing irrational demands on the manufacturers in economic or technological terms? Indeed, what are they? Are they measured sensibly, and do they reflect typical driving patterns and driver characteristics? etc. The Department of Transportation has done very little in public education, perhaps in the fear of being accused of trying to shape public opinion. Even in the very popular courses that engineers offer for liberal arts students at my own institution, we find it very diffi- cult to include transportation issues because there is so little source material available (except in cases such as the Bay Area Rapid Transit history, where we suffer from overdocumentation). Education of the public to the point at which intelligent decisions are possible is the ideal area for government-industry cooperation with and utilization of the colleges. With an appropriate informational effort on a national scale, we can avoid repeating the problems and "disasters" of the l970s in the coming decade. l02

ECONOMIC INCENTIVES TO INNOVATION IN TRANSPORTATION REMARKS BY BRUCE S. OLD This panel, while it is involved in economic incentives, has one area that is forbidden to us, and that is the area of federal R§D. Whereas it is proper to say that federal R§D invested in a certain field indeed represents an economic incentive, that field happens to be the purview of another panel. Now, the fact that we do have certain territorial divisions here brings out the importance of the final panel reports, where, the panel chairmen will be forced to bring about some cross-fertilization among the five panels, because there is a fair amount of overlap. In order that the reader can better visualize the sorts of things we will be discussing, I would just like to mention the skills in the membership of our panel. We include in our membership people who under- stand the motor vehicle industry, the railroad industry, the airline industry, and the freight service industry. We also have three experts from the Department of the Treasury who understand economics and tax policies. We also have an investment analyst who understands the opinion of the investment community with respect to innovation in the transportation industry. A background paper has been prepared for our panel by Aaron Gellman. That paper will be presented by Ed Haefele of the University of Pennsyl- vania. Finally, I would like to make two brief remarks about aspects of the conference that particularly interest me at the moment. First, Court Perkins, the president of the National Academy of Engineering, has said that there had been a dozen studies on innovation, but never any action taken on the recommendations of the studies. It is important that we develop some ideas as to who is on the other end of the telephone, who the different people are we have to be able to contact in the various agencies, departments, and committees of Congress, to assure that actions are taken on our recommendations. Each panel will have different people at the other end of the telephone line. We have to develop a clear idea about who we have to speak to in order to achieve any actions on recommendations. The second aspect has to do with the reality of any economic in- centives that do indeed affect innovation. I have heard for a good many years that it would be nice to change this or that tax policy and l03

then things would happen. I think, however, that if one talks in the abstract, it is very difficult to get anybody excited. I recently ran into a real case history that I think is very excit- ing. On the north shore of Lake Erie, the Steel Company of Canada (STELCO), a steel company operating in Canada, is planning a Greenfield steel plant. On the south shore of Lake Erie, the United States Steel Corporation, which is seven times the size of STELCO, has had a license granted it to build a Greenfield steel plant near Conneaut, Ohio. The Canadians are moving ahead with their Greenfield steel plant now. It is under construction. On the other hand, the United States Steel Corporation Executive Committee is still studying the problem, and the problem there is twofold. It is not just a matter of capital avail- ability; the primary problem is return on investment in this particular endeavor. Now, the Canadians are pulling ahead for a very interesting reason, and it is very simple. They have stated it clearly to us. It is the difference in depreciation allowances between Canada and the United States. The Canadians are allowed to depreciate their entire plant in two years, 50 percent the first year, and they are able to set their own schedule for what they will do in terms of the time taken to depre- ciate the other 50 percent. Furthermore, the depreciation begins the moment equipment is delivered at the plant site. They do not have to have the equipment in operation. Therefore they are depreciating that plant right now, and the United States Steel Corporation has not yet been able to come to a decision. So here is a real example of how differences in tax policies indeed create an ability to innovate. INCENTIVES TO INNOVATION IN THE TRANSPORTATION SECTOR BY AARON J. GELLMAN PRESENTER: EDWIN HAEFELE SCOPE The present paper considers the process of innovation in the transporta- tion sector over less than the full spectrum of "transportation." Specifically, the paper concerns itself with transport that is actually or potentially produced on a commercial scale; both passenger and freight transportation are considered, with emphasis on intercity transporta- tion even though there will be reference to urban transportation as well. The initial mandate for the paper called for consideration only of "economic incentives." It soon became obvious that the scope had to be expanded in two significant ways in order to achieve its overall objec- tives. Specifically, the adjective "economic" had to be dropped, first, l04

because of the difficulty in distinguishing economic from noneconomic incentives in many cases and, second, because what is clearly an "economic" incentive for one party to a given innovation process may be seen in quite another light by others. The second change in scope has been to include disincentives to innovation as well as incentives, since the former have equal or even greater importance, especially where a public-interest sector of the economy (such as transportation) is concerned. Disincentives also need to be considered explicitly if discussion of government's role in im- proving the innovative performance of such sectors of the economy is to be catalyzed. Both public and private policies and practices that tend either to promote or to thwart innovation will be analyzed in the material to follow. Special emphasis is laid upon those incentives and disincen- tives that promote net beneficial innovation in the transportation field, but those which (happily) discourage net costly innovation are not ig- nored. In terms of the elements of society that will be considered either directly or indirectly, there is very broad scope indeed. This, of course, grows out of the fact that given processes of innovation can involve either a narrow or wide range of institutions and individuals depending on the nature of the former and the markets that they are intended to address. In any event, the paper will variously consider any or all of the following as is appropriate: 0 carriers o suppliers to carriers o labor o shippers o receivers o travelers o the public at large o special interest groups 0 government entities at various levels and in various branches Finally, as to scope, the paper refers not only to organizations and institutions, but to individuals as well. At the outset the pro- cess of innovation must be recognized as a "people process." Fundamen- tally, it is people who promote or thwart innovation processes, in transportation as elsewhere. Indeed, individuals play a far more critical role in the process of innovation than is often recognized. So it is that the role of the individual is given substantial emphasis in this paper, without unduly downplaying institutions and organiza- tions, both public and private. DEFINITIONS Some pertinent definitions follow: Invention—to conceive the idea. Innovation—to use the process by which an invention or idea is translated into a product or process and brought into the marketplace. l05

"Beneficial" innovation—an innovation that generates benefits net of costs when recorded in either private or social terms, or both. Public enterprise—industrial or service activity with an identi- fiable output (product or service) owned by govern- ment such as a federal agency, city, or port authority. Shop rights— the right of an employer to use without payment of any royalty his employee's invention developed in the course of his employment. Such use is restrict- ed to the employer's purposes with the employee otherwise free to exploit his idea or invention for his own gain. Perks—fringe benefits, expecially nonmonetary, that form a part of an employee's compensation package. INCENTIVES AND DISINCENTIVES IN THE PROCESS OF INNOVATION The incentives and disincentives that play upon the process of innova- tion are many and varied. This is true whether the transport sector or some other element of the economy is the subject of study. Still, there is probably no area of the economy with a wider range of incentives and disincentives to innovate than transportation. This is due to the ubiquity, the wide range of technology, the complex institutional arrangements, and the governmental involvement that characterize the transportation field. Tables l through l0 provide a comprehensive listing of the incen- tives and disincentives that might be present in any given process of innovation, be it technological innovation or otherwise. The tables have been developed not just with transportation in mind. As was previously noted, however, at some point in the development of the trans- portation system of the United States each one of these incentives and disincentives has been influencial. With regard to Tables l through l0, the odd-numbered tables are devoted to incentives to innovation and the even tables to disincentives. Both incentives and disincentives are listed in terms of the "parties" upon which they bear—parties that can be influencial in shaping and pacing specific processes of innovation. These parties to the process of innovation include the following: o individual persons o individual firms o public enterprises o industries e nations as a whole It should be noted that the incentives and disincentives on these tables are not mutually exclusive. More than one incentive or dis- incentive can be effective contemporaneously in the same process of innovation and can bear upon the same individual or enterprise or nation. Not only that, but some of the incentives (and disincentives) l06

TABLE l Incentives to Innovation That Influence the Individual l. Increased current income 2. Increased future income 3. Nonsalary "perks" of value (e.g., stock options, profes- sional travel) 4. Job promotion or heightened probability of promotion 5. Increased prestige and/or responsibility 6. Job offers 7. Shop rights 8. Opportunity to participate in the application of one's own ideas or invention TABLE 2 Disincentives to Innovation That Influence the Individual l. Lack of rewards, even if "successful" 2. Increased visability 3. Increased responsibility 4. Extra effort required to perfect the "innovation" 5. Likelihood of job change (e.g., new responsibilities and/ or geographical shift) 6. Frustration (e.g., inability to advance a "good idea") 7. Risk of failure 8. Employer attitude toward failure of an innovation process l07

TABLE 3 Incentives to Innovation That Influence the Firm l. Increased current earnings 2. Increased future earnings 3. Achievement of revenue growth objectives 4. Achievement of profit objectives (e.g., reduce costs, stimulate demand) 5. Achievement of corporate diversification objectives 6. Increased market share 7. Increased multiple on stock 8. Capital conservation (e.g., promote non-capital-inten- sive production methods) 9. Reduced dependence on labor l0. Availability of IR§D funds ll. Meet regulatory requirements l2. Presence of regulation that heightens the probability and/or profitability of successful innovation l3. Improve recruitment results l4. Enhanced image l08

TABLE 4 Discentives to Innovation That Influence the Firm l. Insufficent competitive spur 2. Risk of capital loss 3. Capital shortage 4. Short-term earnings penalty 5. Insufficient period of "monopoly profits," even if successful 6. Sufficiently high returns and growth rates without assuming the risk of innovation 7. Durability of capital equipment on hand 8. Inelastic demand for current product(s) or service(s) 9. Rate-of-return regulation employing a deferred rate-base calculation l0. Technological integration (e.g., "lumpiness" of invest- ment need to fit into technologically complex system) ll. Regulation—economic or other l2. Antitrust implication of innovation l3. Industrial standardization (externally or internally imposed) l4. Lack of corporate/divisional growth objectives l5. Risk or fear of "failure" l6. Inappropriate reward structure to promote innovation l09

TABLE 5 Incentives to Innovation That Influence the Public Enterprise 1 . Increased revenues 2. Expanded responsibilities (e.g., functionally, geographically) 3. Increased return on invested capital 4. Improved ratings of debt instruments 5. Amelioration of complaints (from customers, citizens) 6. Meet regulatory requirements 7. Accommodate "customer" innovation 8. Accommodate political pressures (e.g., demand for increased labor intensity in operations) 9. Enhance "owner's" image generally, in the community served and beyond) TABLE 6 Disincentives to Innovation That Influence the Public Enterprise l . Lack of competitive spur 2. Capital constraints 3. Durability of capital equipment on hand 4. Inelastic demand function 5. Absence of life cycle costing 6. Absence of explicit growth objectives 7. Absence of conventional profit-and-loss statement and balance sheet 8. Increased operating costs 9. Lower productivity, labor and/or capital 10. Innovation not required by regulation ll. "Customer" resistance to change l2. Labor content "requirements" l3. Inappropriate reward structure to promote innovation ______ 14. Threat to "1 nw-prnfi 1 e" ll0

reinforce one another. For example, with regard to the incentives fac- ing an enterprise or an industry, increased current earnings or pros- pects of increased future earnings would tend to improve the view investors take of their securities and thus, among other things, increase the multiple of earnings applied to their shares, if they are publicly traded. Not all incentives or disincentives are applicable to every in- dividual or enterprise. For example, by particularizing to transporta- tion Tables 3 and 4, which deal with incentives and disincentives, respectively, with regard to individual private sector firms, it can be observed that some of the incentives and disincentives apply with different force to carriers as compared with suppliers to such carriers of equipment or infrastructural components or services. With reference to Table 4, the application of rate-of-return regulation, which employs a rate base formula including only investments actually in use as con- trasted with those represented by construction work in progress, generates significant incentives to innovate for carriers regulated in this way. Suppliers to carriers are not subject to rate-of-return "rate" regulation in the United States, and consequently this disincen- tive is not applicable to them. If the more or less comprehensive list of incentives and disincen- tives to innovation provided in Tables l through l0 merely conveys the complexity and wide variety of possible incentive or disincentive struc- tures at work where the process of innovation is concerned, it will have materially contributed to setting the stage for the discussion that follows. Certainly in the course of the remaining portion of this paper, only a limited number of specific incentives and disincen- tives will be considered from the total presented. Before considering specific public policies and their relation- ships to the innovative performance of the transportation sector, it should be noted that the presentation of the incentives and disincen- tives has employed a "bottoms-up" approach. This is to underscore the primacy of the individual where innovation process and performance are concerned. No matter what the setting for innovation, there must be one or more individuals who stick their necks out and champion the application of the idea of invention that only becomes an innovation when there is market introduction or a product or service through an arms-length transaction. If individuals are not properly motivated to assume the risks associated with innovation—such risks being monetary, professional, and often very personal—there will be little or no inno- vation. The reasons it is difficult to separate "economic" incentives from other incentives can be seen by studying Tables l through l0. Consider Table l; obviously, increased current and future income clearly repre- sents economic incentives to spur an individual to participate enthu- siastically in an innovation process. On the other hand, increased prestige or visibility for the individual may not result in economic rewards although it is difficult to say one way or the other in advance. But since such increased prestige or visibility may be of "value" to the individual, it is best to consider them under the rubric "economic lll

TABLE 7 Incentives to Innovation That Influence an Industry l. Increased current earnings 2. Increased future earnings 3. Improve financeability 4. Increase share of GNP 5. Thwart foreign competition 6. Promote favorable government action 7. Increased tolerance of industry-wide cooperation 8. Increased visibility (favorable); improved image 9. Improved recruiting results l0. Meet regulatory mandate i TABLE 8 Disincentives to Innovation That Influence an Industry l l. Lack of sufficient competitive spur (high concentration ratio?) i 2. Capital constraints 3. Durability of capital equipment • 4. Technological integration 5. Standardization (externally or internally imposed) i 6. Inelastic demand for industry output 7. Regulation—economic or other; regulatory process 8. Rate of return regulation and deferred rate base ] calculation 9. Fear of hurting weak competitor (especially in highly concentrated industry ll2

TABLE 9 Incentives to Innovation That Influence the Nation l. Increased GNP (real) 2. Enhanced productivity—any and all factors 3. Increased employment 4. Improved distribution of income 5. Increased development of new enterprises 6. Improved U.S. balance of payments: cut imports/expand exports 7. Improved "quality of life" 8. Increased decentralization of industry 9. Enhanced international prestige l0. Strengthened military posture TABLE l0 Disincentives to Innovation That Influence the Nation l. Regulation—economic, safety, environmental; regulatory process 2. Chronic inflation 3. Tax level and structure 4. Egalitarian philosophy (e.g., redistribution of income objectives) 5. Decreased employment (e.g., from automation) 6. Natural resource constraints 7. Import barriers 8. Export barriers 9. Golden Fleece-type awards ll3

incentive." To be sure, history records myriad cases, in transportation and elsewhere, where individual star performers in the process of inno- vation have been motivated at least as much by nonmonetary gains as by other factors, but the line between economic and noneconomic is too fine to draw. Besides, doing so would seriously distort the discussion and likely lead to inappropriate conclusions in some cases. The same point can be made in the context of the public enterprise, as is shown on Table 5. Here some of the incentives are clarly "econo- mic," such as the desire to increase the revenues of the public enter- prise (e.g.? commercial airport, municipal transit company, port authority), but is it a purely "economic" incentive that successful innovation could lead to the expansion of the responsibilities of a specific public enterprise? First of all, it might be an economic in- centive if increased salaries or other rewards were thereby made availa- ble to the proprietors of such a public enterprise, but if the only incentive was the enhancement of their political leverage and prestige and if that was sufficient to spur them to sponsor and support innova- tive activity, it should certainly be considered, since one cannot be sure that each and every vector of each and every incentive or dis- incentive eventually will not have economic consequences or implica- tions. PUBLIC POLICIES THAT INFLUENCE TRANSPORT INNOVATION In part because of the ubiquity of transportation, literally every public policy in some way influences the process of innovation in transport. Nevertheless, there are several specific but broad policy concerns of government that exercise particular and continuing leverage upon transport innovation, and some of these will be the focus of the present discussion. To be considered are selected policies related to the following: & competition o the purchasing function 0 financing 0 public enterprise o antitrust 0 market aggregation 0 identification and amelioration of social or external costs Where appropriate, the opportunity will also be taken to consider various ways in which public and private policy interact to produce incentives and disincentives to innovation in the transportation field. Competition Perhaps the most fundamental and long-standing concern of government where transport is concerned relates to the level and character of competition that obtains in various provinces of the transport sector. Each and every legislative mandate handed to a regulatory agency by ll4

the Congress (or by a state legislature) has cited the preservation, control, or partial elimination of competition as an objective of the legislation and of the regulation that it establishes. From Tables l through l0 it can readily be seen that competition, either explicitly or implicitly, represents an important incentive or disincentive to innovation. Not only must government's concern with competition in transportation be related to the innovative performance of those producing and marketing transportation services, but it is necessary to consider the influence of public policies toward competi- tion on enterprises that supply producers of transportation with the inputs they employ. During much of the time that transport economic regulatory agencies have been operating in the United States, such agencies have reflected a conservative attitude toward those segments of transportation that are of direct concern to them. Specifically, entry/exit and pricing regulations have usually been structured largely to preserve the compe- titive status quo. That is, entry and exit have both been substantially constrained in the principal means of transportation in the United States and prices (i.e., rates and charges) have also been subject to such constraints that they have been far more rigid than would have been the case without the regulation. Under such conditions, which often include the imposition of price identity among competitors in the same market, it is not surprising that the propensities to innovate of the regulated (and their suppliers) have been dramatically different than would have been observed without such regulation. One classic illustrative case concerns the U.S. trunk airlines in the period between the end of World War II and l960. In this period, under the rate regulatory scheme applied, an airline often put great stress on having flight equipment that was technologically different—advertisably different—from that of its competitors. The stress on establishing such a difference was so great that unit produc- tion costs (e.g., cost per available passenger mile) were often a secondary consideration in the choice of new aircraft by U.S. scheduled air carriers. The incentives of increased market share and heightened rate of growth caused many airlines ultimately to choose flight equip- ment that was more expensive to operate on any basis of calculation than was other equipment then available. Both the Lockheed Constellation aircraft powered with turbo-compound engines and the DC-7 owed their existence to this situation. Fortunately, the regulatory schemes applied in several areas of transportation either have been changed dramatically or are in the pro- cess of change. Specific reference can be made to revision of the economic regulation of U.S. commercial air transportation that calls for all but complete abolition of industry-wide price guidelines by the mid-l980s. "Deregulation" is also being discussed in the context of highway and rail transportation, with one of the explicit objectives of such "deregulation" (more accurately termed "regulatory reform") being to place the incentives of cost reduction and demand stimulation in proper persepctive as far as firms and industries are concerned. More enlightened (or less) economic regulation would have led to a very ll5

different history with regard to the Douglas DC-7 and the later versions of the Lockheed Constellation. Here the airlines eschewed cost reduc- tion as an incentive to innovation and embraced totally the incentive of demand-stimulation. At least a better balancing of such motives would have resulted under a more enlightened approach to the regulation of competition in air transportation. The best public policy toward transport (as in other fields) would seem to be one that encourages entrepreneurs to place the proper weight on these most fundamental incentives to innovation, cost reduction and demand stimulation. Where competition is overly constrained by regula- tion, the power of these incentives are distorted both in absolute terms and in relation to each other. This suggest, of course, that whenever public policy toward competition in transportation is being considered either de novo or on a review basis, there should be explicit analysis of the impact of such regulation or regulatory change upon the several processes of innovation that will be influenced by such compe- tition as may be created or discouraged. The regulation of competition in the transport sector has usually been such as to preserve substantially the status quo not only intra- modally but intermodally. An excellent case in point concerns a sub- stantial railroad technological innovation of the l960s, the Southern Railway's "Big John" covered hopper grain cars. As part of a well-thought out program to enhance both its traffic and its profits, primarily at the expense of inland waterway carriers, the Southern embarked upon a program to increase materially its share of the market in grain traffic between the Midwest and the Southeast. A central part of its strategy was the employment of new and highly innovative aluminum-covered hopper cars with a nominal capacity ot l00 tons. The incentive to the Southern was clearly to expand markedly its grain traffic by reducing rates—but only where such rate reductions were more than justified by the cost reductions that would be experienc- ed through use of the Big John cars in multiple-car and unit train services. In short, the cost reductions would support rate reductions that in turn would shift the demand from the barge to the rail mode of transportation. (The Southern also expected to gain traffic at the expense of other competitive rail carriers, but this was a relatively small portion of the traffic gain they expected to enjoy in the long run.) Using its rate-regulatory powers, the Interstate Commerce Commis- sion (ICC) delayed the Southern's introduction of the new and lower rates, and between the ICC and the federal court system, the Southern was years in realizing the full benefits for itself (and for the ship- pers and receivers of grain) that it had projected and that were sub- stantially proved to be "deliverable" once the railroad became able to use Big John as had been intended from the beginning. To achieve its objectives with Big John, the costs imposed upon the Southern by the regulatory scheme of the ICC were very high, in terms of both direct expenses and opportunity costs. While there has never been an explicit accounting from the Southern, it is estimated that the costs related to the workings of the regulatory scheme were ll6

in the range of $20 to $40 million in the case of Big John. The point, of course, is that a regulatory approach, such as has been imposed traditionally by the ICC, significantly undermines such incentives to innovation as are shown in Table 3 in items l, 2, 3, 4, 6, 7, 9, and l4. Moreover, with reference to Table 4, which deals with disincentives to innovation that influence the firm, such regulation certainly heightens the disincentives labeled 2, 4, 5, ll, l5, and l6. But intramodal and intermodal competition in transportation markets is not the only concern of public policies toward competition, which, in turn, work on the incentive structure that is relevant to the innova- tion performance of the transportation sector. Both the framers and the administrators of economic regulation in transportation more often than not overlook the implications of such regulations on innovation in those industries that supply carriers with the equipment, infrastructure, and services that such carriers organize so as to produce transportation. Returning to the case of the DC-7 and advanced Super Constellations, for example, it has been demonstrated that the system of price-identity regulation in the airline field was as responsible as any other single factor for the competition among manufacturers of aircraft to develop relatively inefficient aircraft that, however, embodied substantial "advertisable" differences when used by regulated air carriers. Thus the propensities and incentives to innovate of both the carriers and their suppliers of airframes and engines were distorted. The relationship between carriers and suppliers and the innovative performance of each can be seen in other contexts as well. For example, consider the railroad supply field. Little concern has been expressed by regulators—and to the ICC must be added the Antitrust Division of the Justice Department and Federal Trade Commission in this connection — as far as competition in the railroad supply field is concerned. For instance, the concentration ratios in several areas of railroad supply have been high for decades and remain high today. And by "high" is meant l00 percent at the four-firm level, which tends to mask the fact that concentration in some fields approaches l00 percent at the two- firm level. The most dramatic cases in point, perhaps, relate to rail- way braking equipment and railway signaling equipment. In each case, just two firms dominate the market. Each enjoys approximately 50 per- cent of the business and supplys technologically nearly identical com- ponents and equipment year after year. Why are the incentives to innovate so weak (or nonexistent) in the railroad braking and signaling equipment fields where both the railroads and the suppliers of such equipment are concerned? First, there has never even been a hint on the part of federal regulators—either ICC or antitrust—that they are concerned with the situation in either market. What makes this malaise especially surprising is that one of the two firms in the braking equipment business, Abex, in the l960s became a wholly owned subsidiary of a major railroad, the Illinois Central, which acquisition, in fact, required the ICC to look into the propriety of the matter and which certainly gave the Department of Justice and the Federal Trade Commission the opportunity to do so. Second, the railroads themselves seem to have little interest in changing the technology employed with regard to the braking of railway ll7

trains and the signaling of railway rights-of-way. This can only be inferred by the fact that railroads continue to publish fundamentally the same design specifications for braking and signaling equipment year after year despite the terrible and growing costs they incur as a result of what constitutes a "no-innovations" policy in each of these areas. Under such circumstances, it is not surprising that the firms involved, Abex and Wabco in braking equipment and Wabco and General Railway Signal (GRS) in the signaling markets, have adopted positions that lead to minimal technological change and then only where produc- tion cost reductions inure largely to the benefit of the manufacturers. Perhaps the most important point is that even when a procompetitive position with regard to transportation companies is maintained, at least at certain times and in certain places, the regulators' ignorance of the situation in the "supply trades" with regard to competition suppresses many of the incentives for innovation as far as suppliers are concerned and heightens the disincentives to innovation in other cases. In terms of Table 3, the situation first described thwarts the incentives to inno- vation designated l, 3, 4, 6, and 7. In Table 4, the disincentives to innovation that gain emphasis include l, 4, 6, 8, l0, l3, and l5. This example suggests that the concept of maintaining effective competition as a spur to innovation in the transportation field must encompass both car- riers and suppliers. And it is not merely the industry-specific regulatory agencies that are involved, but also the antitrust "watchdogs" such as the Department of Justice and Federal Trade Commission at "the federal level. Government concern with the suppliers of transportation equipment, infrastructure, and services, is not and cannot be confined to govern- ment's role as a regulator of competition. The government is itself a major factor in the acquisition of transportation services, and, in some instances, purchases transport equipment and infrastructural com- ponents as well. It is reasonable that supplier performance, and public policy towards the purchase of transport equipment, infrastructure, and services be of interest to transportation enterprises and the government. Purchasing What is sometimes referred to as the "new golden rule" holds that "he who has the gold makes the rules." There clearly is truth in this, and it is especially relevant in the context of innovation where parties that have a manifest demand for goods or services should have quite a lot to say about the character of the products or services that they require--provided they care about it enough in the first place to exert such influence. It is power from the demand side that is often a major catalyst to innovation, just as, in other instances, there is a greater measure of "supply-push" than "demand-pull." Without doubt, however, the way in which the purchasing function is exercised can have a profound influence on the process of innovation in that it can heighten or suppress both incentives and disincentives to innovation. ll8

Public policy toward purchasing in the transportation field can be expressed most directly through the power for the purse that the federal government can reflect by its promulgation of "rules" under which its money will be spent. The public sector, in the aggregate, remains the largest single purchaser of intercity freight transport services in the United States. Yet, it has not often exercised this power to "force" private entrepreneurs to furnish transportation services that, in turn, require such carriers to exploit technological possibilities beyond those already underlying the transportation "end product" being offered and used. There would appear to be some considerable leverage available to the federal establishment in its role as shipper; judicious applica- tion of such power to induce net beneficial technological innovation in the transport sector could not help but to be a laudable exercise of public policy and power. Public policy translated through the purchasing function is mani- fest in a number of other ways as well. For example, the federal government often provides enterprises with the major portion of the financing necessary to acquire given pieces of transportation equip- ment. An especially interesting case in point relates to the procure- ment of buses for urban transportation systems, where the federal govern- ment has, in fact, attempted to exert considerble influence on techno- logical innovation. The program through which this is most clearly demonstrated is called Transbus. With Transbus, it is important to keep in mind that the federal government was providing funds necessary for the development of the vehicle but also was committed to supply the greatest proportion of the capital needed for its ultimate procurement by transit firms. Moreover, virtually all the transit companies that would ultimately acquire Transbus would themselves be public enterprises, rather than private, and would also be obtaining on a continuing basis a major share of their operating expenses directly from the federal government, from the same agency that was funding Transbus development and that would fund Transbus procurement. In such a setting, one would normally expect the federal government to have considerable leverage on the process of innovation. And, indeed, it did. As Transbus emerged more clearly in terms of its role in society under its function as an item of transportation hardware, it became apparent that significant technological innovation would need to be incorporated in the vehicle if it were to meet all its objectives. Certainly, the multitude and variety of objectives loaded on the back of Transbus in and of itself represented one of the fundamental influ- encing factors where innovation was concerned. Moreover, some of the goals sought to be achieved through Transbus were clearly conflicting. For example, the requirement that Transbus readily accommodate elderly and handicapped (E§H) travelers necessarily made the vehicle substanti- ally heavier than it would otherwise have been; at the same time, maximum energy efficiency was also an explicit mandate. In the face of such conflicts, actual and potential manufacturers of Transbus, who had to be innovative if they were to succeed to an order, must have been confused as to which goals took priority since not all of them could be achieved simulataneously. ll9

To make matters worse from the standpoint of the process of innova- tion, the federal government in its role as midwife to Transbus (a mid- wifery wholly financed by the federal government), after almost a decade and after direct expenditures in excess of $25 million, came up with what was very much a design specification and very little a performance one. Since one of the great catalysts to innovation in a market such as that for public transit vehicles is the true performance specifica- tion issued to any and all parties interested in participating in the project, the emergence of what was essentially a design specification in and of itself severely thwarted the process of innovation and, ultimately, the Transbus project itself. Coupled with the reliance upon a specification couched largely in design rather than performance terms is the fact that the attempted first procurement of Transbus, which required the aggregation of demand of three major metropolitan areas in the United States (Miami, Philadel- phia, and Los Angeles) absolutely ruled out the competitive responses being judged in terms of life cycle costing for the Transbuses to be procured. In fact, the Request For Quotation (RFQ) went to the other extreme and made initial capital cost the sole basis for determining the winning bidder. Once more, the abject ignorance of the process of technological innovation manifest through such a policy suggests that perhaps one of the great incentives to beneficial technological innova- tion in transportation would be to assure that those public officials establishing the policies and rules associated with the procurement of innovative products (or services) know a great deal more about the pro- cess of innovation, and particularly about the private sector's invest- ment decision-making processes, than is now obviously the case, at least as reflected in the Transbus program from start to finish. The necessity of relying upon true performance specifications if the process of technological innovation is to be made more effective in the transportation field cannot be overstressed. Not only is this important where transportation equipment is to be procured, but also where infrastructural components are involved. Consider the case of the reconstruction of the Northeast Corridor railroad network. With funding coming almost entirely from the United States Treasury, as administered by DOT, the specification employed to procure the signal- ing system for the corridor was so much a design specification that, in fact, only two qualified bidders emerged, the railroad industry's old "friends," Wabco and GRS. So it is that a substantial number of potential competitors, some domestic and some foreign, were effectively ruled off the track before the race even began. Also a situation has been produced in which, in the l980s, a signaling system featuring mechanical relays and vacuum tubes will be installed in the Northeast Corridor. Not only does this adversely affect the future economic and operational performance of the corridor, but it also denies the railroad industry as a whole the benefits of dramatic technological innovations that unquestionably could have been induced through the corridor pur- chasing power had the right form of specification been employed in the procurement. It is worth pointing out that in many transportation programs where federal funding is heavily involved, including Transbus and the l20

resignaling of the Northeast Corridor, the issue of "buy American" inevitably arises. And yet, the effect of the bias on the part of the United States government, not surprisingly, has been to handicap, if not totally thwart, foreign suppliers in such competitions. While the political basis for "buy American" is clear and understandable, its effects on the process of innovation (and especially upon the presence or absence of a competitive spur in the marketplace) are neither as clear nor as defensible. In the case of Transbus, it was only several months before the bid opening date for the first aggregated purchase of Transbus that the secretary of transportation indicated he might permit foreign suppliers to participate. This was one of the elements leading to the situation that when the bids were to be opened there were none—either from domestic or from non-U.S. suppliers. Certainly competition is a tremendous incentive to innovation, as has previously been established, and public policy ought not blindly rule out foreign production possibilities as a source of generating such competition, particularly where innovation is urgently required to achieve various social, political, and economic goals and where the expected U.S. respondents are few, such as in both the urban bus and the railway signaling fields. Another major point needs to be made with regard to purchasing policy and its influence on innovation in transportation. There are agencies of government that are the primary source of demand--even the only source of demand—for certain hardware and software related to the production of transportation services. An especially good case in point is the Federal Aviation Administration (FAA), which specifies, designs to a great extent, procures, owns, and operates several systems essential to the production and growth of air transportation services in the United States. This is a case, not unprecedented, where an agency is both the judge and the jury with regard to the technology and technique being employed. Under such circumstances, it would seem prudent that the agency bend over backward to induce competition in every possible dimension among the suppliers upon which it must rely to produce the hardware and software that are at the heart of the sys- tems which it operates. But this is not what happens. Instead, at the very beginning of the process of innovation, where creativity is at a premium, the En- gineering and Development (E§D) elements of the FAA do not systematical- ly employ techniques of procurement or of publicity to induce imagina- tive external responses to the needs of the agency with regard, for example, to its air traffic control systems. In fact, FAA E§D person- nel have often indicated over a considerable period of time that what the FAA requires in the way of equipment and infrastructure and soft- ware is not likely to be supplied by small enterprises, and so they justify the concentration of their procurement activities and communica- tions with quite large enterprises that just happen, in most cases, to be the suppliers they have done business with before. Obviously, this flies in the face of myriad analyses that support the hypothesis that it is small enterprise that is most creative and innovative and that a very substantial proportion of the genius of this country for l2l

reaching to the frontiers of technological possibility is lodged in small enterprise rather than in large. Once more, part of the problem can be traced to the overreliance of the FAA on design specifications— such designs having been developed either by FAA personnel in-house or by firms under contract to FAA. Most thwarting for a large segment of the population of innovative enterprises, however, must be the abject ignorance of the nature of the process of innovation on the part of most persons in the FAA, whether they are involved in research and development, procurement, installation, or maintenance. Finally with regard to purchasing policy, a subtle issue of con- siderable importance to the innovative performance of the transporta- tion sector has long been overlooked. In the United States, there has sprung up over the years a substantial number of "small R§D firms" that are highly creative and are capable of carrying out research and develop- ment for their customers in an efficient and timely manner. A very large proportion of these firms, however, work exclusively for the federal establishment and have long since defined their goal as making a profit through the performance of research and development contracts. These firms have no commitment to carry the results of their R§D further into the marketplace so that the process of innovation can be completed. In contrast, their research and development results are couched in terms that government executives will understand rather than in terms that promote the onward exploitation of their outcomes. So it is that there are myriad R§D results on the shelf throughout the federal government, including the transportation agencies, that will never be exploited through the process of innovation in the marketplace. There arises a policy issue as to whether the federal government ought to continue relying upon such enterprises in which the sole reason for existence is to produce R§D results, with the "small R§D firm" never it- self becoming committed to carrying such results forward to the "real world." There is certainly need for data on the subject, but at this point it would seem that the process of innovation in transportation would be substantially improved if the government, through its purchas- ing of R§D results related to hardware and software (as opposed to policy research, for example) were to encourage the onward exploitation of "successful" R§D outcomes rather than discourage them by allowing, if not requiring, a break between the completion of the R§D phase of an innovation process and the commencement of the technology delivery phase. One of the ways to do this is to restrict severely the reliance of trans- portation agencies on contract R§D firms for R§D results. Financing The availability of financing is often an incentive to the process of innovation; a lack of available financing is always a disincentive. Consequently, the public sector has substantial leverage it can bring to bear on the process of innovation in transportation in general, as well as on specific innovation processes, through the granting or with- holding of financial support as may be consistent with public policy. l22

Financing is an issue in several stages of the process of inno- vation. For example, the financing of basic research (and of applied R§D in many cases) presents government (and private industry as well) with issues that are very different from those presented by financing to support the diffusion of the results of technological innovation that have already been demonstrated to be practical. Yet financing from the public sector has an important role to play in each of these cases and in many others besides. The activities that lie at the "front end" of innovation, i.e., the research and development sort of activities, contrast sharply with those associated with market diffusion (e.g., the purchase of a national fleet of Transbus equipment). In the former, there is much speculation associated with the investment of resources. (And the more "research" rather than "development" the nature of the activity is, the more un- certain the outcome.) It is in the earliest stages of the process of innovation, then, that private sector entrepreneurs tend to be most reluctant to commit their own resources. Consequently, government financial support is most necessary where R§D is concerned. Further along in an innovation process, external financial subsidies required by the private sector or by public enterprises may be materially less, at least in terms of the proportion of resources required from govern- ment. This is because the risk is presumably reduced as the innovation process proceeds in the direction of the market (or else the particular innovative activity would have been killed) and because the time between investment and payoff is much reduced in comparison with the situation where the earliest innovation process activities are concerned. Still, in many cases public support is absolutely required if the process of innovation is to be completed and market diffusion of products or services generate social benefits in excess of external costs, which is often the case in the transportation field. Again, the Transbus program represents a case in point, which is especially apt since the potential acquirers of Transbus were anticipated to be public enterprises. Public financial support for innovation can be introduced in myriad ways. Perhaps the most direct method is through grants for the acquisi- tion of equipment or infrastructural components that are conditioned by the requirement that the funds cannot be used unless some measurable quantum of innovation is reflected in the material acquired with the grant money. This can be effective, when intelligently administered, but it can also produce little but confusion and waste (as in Transbus to date). As was noted earlier, a large and disproportionate share of inno- vative activity, expecially at the front end of the innovation process, is lodged in smaller enterprises. Consider also that one more or less common thread in the development history of new items of transportation hardware and software relates to the terrible expense of prototyping and testing the results of a "successful" R§D effort. This suggests that federal financial support might well be made available to promote the process of innovation in the transportation field by making it possible for entrepreneurs—especially in small enterprises—to obtain prototyping and testing "services" at a cost they can afford—at a cost l23

sufficiently low that these unavoidable steps in the process of innova- tion can be traversed with minimum delay. Providing "in-kind" support in many cases is just as effective in the process of innovation as pro- viding cash, and perhaps even more so. (It is also worth noting that in some instances the prototype and test facilities to support transpor- tation innovation are tha same sort of facilities the military establish- ment requires for similar purposes. In such cases, the joint use of existing public investments might be possible, thus reducing the direct financial burden on the Treasury while improving the innovation per- formance of the transportation sector in a measurable way.) To return to the process of innovation occurring in the market rather than beginning in R§D, it is obvious that much innovation is denied because manfacturers of hardware and infrastructural components see a history of feast or famine with regard to demand even where the results of the innovation process are highly beneficial and attractive to those who must make a purchase decision in favor of the innovation. In many cases, a long history of feast or famine on the demand side has been sufficient to discourage the allocation of private resources to the innovation processes through which transport equipment and infra- structure components would be upgraded technologically. There are no data or information to support an intelligent hypothesis about the leverage—the negative leverage—that the not uncommon feast or famine character of demand has exerted on the process of innovation in trans- portation, but it is a reasonable speculation that it has been consider- able. (Still, one must always remain alert to distinguishing between excuses and reasons and not be overly discouraged when supposed barriers to the process of innovation are dismantled only to find that the influ- ence on the process of innovation had been minimal or even nil. At least some excuses will have been removed from the scene.) In areas where there is a history of feast-or-famine demand, it is entirely possible that one of the most effective ways to employ federal financial resources to improve the process of innovation is to smooth out the demand for hardware and infrastructural components. Perhaps the government can guarantee a certain minimal demand so that the difference between lean years and fat years is not so dramatic as has often been the case, with railway freight cars, for example. If, but only if, such a program were judiciously and rigidly administered, it could generate mutually beneficial results for the transport sector and for the economy; it requires only that the public sector invest capital for various periods of time as would be associated with the stockpiling of output during periods of slack demand and the distribution of such output as had been stockpiled in periods of peak demand. Once demand is stabili- zed, technologically improved output can be expected whether or not public financial support was coupled with an explicit requirement that manufacturers extend themselves in this direction. In many cases—though certainly not all—transport "delivers" to society external benefits in excess of external costs. Where this is so, it becomes rational public policy to support processes of innova- tion that improve transport sector performance. In this connection, suggestions have been advanced periodically over the years that the l24

federal government ought to establish a "transportation equipment development bank" that would provide present and prospective producers of transportation equipment with a source of low-interest, long-term financing explicitly to support the process of innovation, whether it be at the front end or in the technology delivery phase of the process. This approach reflects the notion that innovation will be more efficient and will be carried out in a more timely manner if federal funding support is reflected on the liability side of the balance sheet of those firms that take advantage of such financing. The "bank" still provides a substantial subsidy given the low-interest and long-term nature of the financing contemplated, but the net financial burden on the public treasury should be tolerably low and more than recompensed by social benefits realized in the long term. One of the most interesting, and not entirely philosophical, financial issues concerns the extent to which federal financial support should be provided to private enterprise as opposed to public enter- prise. Specifically, is it wholly rational (economically) that public funds for the support of the "aviation system" be allocated to public enterprises (such as airlines)? Clearly, there comes a time when the marginal utility of a dollar is less to the former than to the latter even when only the social-benefits-to-social-costs relationships are assessed. Yet there is little clamor when support is initiated or is increased for public enterprises, even nonfederal public enterprises. In contrast, great objections are usually raised when financial support to the process of innovation is suggested for private entrepreneurs in transportation. It would seem appropriate that this issue, and the determination of the appropriate balance, mode by mode, ought to be the subject of continuing analysis on the part of DOT which, in turn, should have the courage of its ultimate convictions as to where such financial resources will do the most good. Public Enterprise The public enterprise is becoming increasingly important in the trans- portation sector. Although a detailed calculation apparently has never been made, it is clear that the share of macroeconomic activity attri- butable to public enterprises in the United States is growing year by year. Moreover, if such a calculation were particularized to transpor- tation, it would also show a rising trend, especially since the costs (and revenues) associated with intermodal terminal operations where public enterprises especially flourish are clearly growing both in absolute and in relative terms. Also, there is an increasing tendency for government in various guises to become involved in transportation in various ways. (Consider the history of transit enterprise in the United States, for example, as well as the recent and growing acquisi- tion of rail properties by state and local governments.) Notwithstanding this growing reliance on public enterprise, little attention has been paid to the management of such acitivites as opposed to the management of private enterprise. Regrettably, even the "better" l25

graduate schools of management in the United States have all but ignored the special problems and opportunities associated with public enterprise, perhaps because academics themselves have not taken the time to study public enterprise management as a field unto itself. A comparison of Tables 5 and 6, which relate to public enterprise, with Tables 3 and 4, which refer to private firms, indicates that the incentives and disincentives at work in the two settings are often very different. In part this grows out of the fact that in the private enterprise there is present the discipline of the profit and loss state- ment and the balance sheet — something substantially lacking in the typical public enterprise. Moreover, the private enterprise will gen- erally be found operating in a competitive setting where to some extent the management can be judged (and can judge itself) through appropriate comparisons with other enterprises operating in the same market. Most often the public enterprise has a relatively strong monopoly position or at least a position that is greatly protected from competition in the marketplace. Consequently, the sorts of performance comparisons available to private entrepreneurs are usually denied the managers of public enterprises. So it is not surprising that the incentives and disincentives are somewhat different and often have different leverage to exert where the process of innovation is concerned. The lack of competition in markets in which the public enterprise is active, coupled with the highly politicized nature of public enter- prise decision-making processes, makes it necessary that external "force" be brought to bear to introduce some surrogate for competition, if only to enable the public to judge the performance of public enter- prise managers against a reasonable standard. Therefore one effective incentive to innovation in the public transportation enterprise may well be the devising and application of means for judging (at least in rela- tive terms) the performance of public enterprise managers in one geo- graphical setting compared with the performance of those in another. This may require that, as a condition of federal funding, a uniform system of accounts for all the public enterprises in a given field be used—a concept public enterprise managers will surely resist with all the vigor they possess. Yet this may also be one of the best means of providing the necessary "incentive" to make public enterprises act more rationally and more aggresively where there are socially acceptable opportunities to innovate or to accept innovation. Of course, one of the major problems with innovation in the public enterprise reflects the reality that the rewards awaiting the successful entrepreneur are usually not nearly sufficient to justify his taking very great career risks through a flirtation with failure, which is always a possibility where significant innovation activities are undertaken. Indeed, the absolute fear of failure may thwart every attempt to upgrade the innovation performance of public enterprise entrepreneurs. While this is a problem not unique to the public enterprises, it i^ true that in the public enterprise, failures are often overstressed and this can- not but reduce the enthusiasm for assuming risks through the sponsor- ship of innovation in such enterprises. l26

J.A. Schumpeter (Harvard) and many other economic theorists have debated the issue of the extent to which enterprises possessing mono- poly power will or will not have a high propensity to innovate. It is reasonably clear that if the monopoly position is one that is so greatly protected that challenge is all but doomed to failure no matter what its source, innovation is not likely to be undertaken with enthusiasm because it is risky and usually requires extra exertion--exertion that managers may very well eschew with impunity, especially in public enter- prises. In the transportation field, such enterprises are certainly in a protected monopoly position and their managers may not happily under- take activities that generate added risk and responsibility. If innova- tion is desirable from public policy and economic performance stand- points, it will usually have to be induced by external forces. One of the ways of providing the incentives required therefore is the identi- fication of the external forces that can be effective and the devising of means for these forces to be applied in the appropriate degree and with the correct timing. Once more, the concept of a uniform system of accounts suggests itself, but the power of demand-pull should also not be ignored. With regard to the latter, it should be noted that where there has been successful and beneficial innovation in public enter- prises, it has often come about through pressures of demand that cannot politically or otherwise be ignored by the public enterprise manager. The periodic revisions observed at most airports represent a par- ticular case in point. Airport management would generally like to live a quiet life, as might be expected, but is often unable to do so because the technology and the patterns of service continue to change where their prime customers, the airlines, are concerned. Yet in order to be continuing institutions, airports must respond to the needs of their most powerful customers, the air carriers. Indeed, there are few cases in the history of airport development in which the lead has been taken by airport management where innovation is concerned. And in those few instances, even when the result turned out to be highly favorable for most, if not all, of the parties concerned (such as at Dulles Interna- tional Airport), the innovative drive of the public enterprise manager was initially resisted by airport users. In a way, it is fortunate that the federal government supplies a substantial proportion of the resources required for most public enter- prises, including airports, and therefore has the ability to impose certain "standards" with regard to the innovation performance of such public enterprises. Unfortunately, the federal government has shown little inclination to condition its grants to public enterprises in such a way as to promote socially and economically beneficial innovations, but this may be because the administrators of such grant programs have themselves too little knowledge and understanding of the form and func- tion and economics of the public enterprises that they are supporting. It should not be difficult to redress this situation, if only the leader- ship in DOT has the will to do so. Once again, an improvement in public enterprise performance, as measured partially by their activities in support of beneficial innovation, should not be difficult to achieve, and the cost to the public in financial terms should be close to zero at the outset, and actual savings should be realized in the long term. l27

Certainly, compared with private enterprise the typical public enterprise has distorted propensities to innovate, given the politicali- zed environment in which the latter operates. For example, it is entire- ly possible that in the present era there is too much stress on energy conservation and noise reduction. By "too much stress" it is suggested that perhaps the benefits gained in terms of energy and noise may be overweighed by the costs experienced by either the public enterprise it- self or those who use it. Operating in a political fishbowl, and with energy and environment concerns rampant, it is difficult to fault the public enterprise manager who is perhaps oversensitive to the times. Still, DOT, in part because of its power of the purse, ought to be in a position to leaven the loaf as necessary. Again, public enterprises often have the wrong incentives in mind when they are making decisions whether to innovate or support innovation. For example, revenue maximization is often the overriding objective of public enterprise managers, given the nature of their financial struc- ture. Of course, students of management have long known that revenue maximization frequently leads to the wrong decision where investment and operating policies are concerned, and this is no less true in the context of the public enterprise than of the private firm. Given the highly personal nature of the process of innovation, it is important to recognize the disincentive to innovation that is pre- sent through the scarcity of cases in which public enterprise managers who have borne risk intelligently through support of innovative activity have as a result achieved visible professional and personal success, however measured. Certainly, DOT can afford to spend the limited resources required to gain a better understanding of what constitutes "success" in the minds of those who are prepared to devote themselves professionally to the management of public enterprises in transportation and subsequently to condition its support of public enterprises partial- ly on the establishment of the preconditions in those public enterprises that reflect the professional and personal needs of such managers. This may well prove to be the most effective single means of improving the innovative performance of public enterprises over the next decade or more. Antitrust It must be recognized that the process of innovation proceeds by fits and starts. It is not a continuous process in any given field, trans- portation included. There are "breakthroughs" followed by long periods of what appear to be technological stagnation. The discontinuous pro- cess that is innovation contrasts sharply with the controlled and evolu- tionary development of the law. When legal considerations and con- straints are imposed upon the discontinuous process of technological innovation, the effect often is to ration technological possibilities to the ultimate marketplace even in times when the latter are being generated at a rapid rate. Consequently, the imposition of antitrust constraints on the process of technological innovation is an especially l28

important issue that has been little considered even in the "regulated industries." One of the "reasons" most often cited by would-be sponsors of inno- vative activity for their lack of enthusiasm for specific innovation pro- jects is fear of antitrust entanglements, especially if successful inno- vation were to result. In many instances, options are absolutely fore- closed to industrial entrepreneurs simply by the uncertainty of antitrust policy toward a specific institutional arrangement that is necessary to induce innovation. A particularly important case in point relates to the development, production, and marketing of transport aircraft at the present time. In order to introduce a new commercial aircraft, even enterprises with the size and character of Boeing or McDonnell Douglas or Lockheed, are required to play "bet-your-company." That is, the resources re- quired to do a new transport exceed in most cases the net worth of the firm. Moreover, given the nature of the technology and of the market for such aircraft, the relationship between the "launching cost" of the new aircraft and the net worth of the firm is becoming increasingly less favorable to a proinnovation decision. Facing such a situation, one of the several options open to airframe manufacturers is the joint enter- prise approach to new projects. But under present conditions, no prudent management of an airframe manufacturer in the U.S. can be ex- pected to explore seriously with one of its historic domestic compe- titors the possibility of a joint venture for the next round of compe- tition where transport aircraft are concerned. This is largely because the Antitrust Division of the Justice Department will give no assurances that such a move would not be viewed as "anticompetitive" and therefore as a step that would be intolerable to the Antitrust Division. Certainly, if one were to define the market for transport aircraft as being the United States alone, this view of the Justice Department might—but only might—have merit. But an increasing proportion of the demand for transport aircraft is found outside the United States. Indeed, over half of the orders for large transport aircraft currently emanate from non-U.S. sources for the first time. If the market for transport aircraft is viewed as a global one, as it properly should be, then an amalgamation or joint venture involving only U.S. firms is not so onerous, expecially given the fact that the joint venture is becom- ing more or less a "standard" means of exploiting technological possi- bilites in other countries (e.g., Airbus Industries). Under the circumstances, in the United States it is increasingly clear that the range of technological possibilites that can be exploit- ed in the form of new transport aircraft is severely limited by the recalcitrance of the Antitrust Division, which refuses to recognize rapidly changing conditions of supply (and of demand) in the transport aircraft field. Indeed, summing up (only a little unfairly) the atti- tude of the Justice Department, their position is that they will react to any proposition that becomes a reality but will not react to a hypo- thetical proposition such as posed by the question, "if Boeing and Lockheed were to form a joint venture, would you intervene?" Without the answer to the theoretical question, the disincentives to certain l29

kinds of innovation in the air transportation field are so great as to be totally thwarting. Yet, antitrust officials cannot, or at least will not, understand this situation; this is not entirely surprising given their background and training in a field that is characterized by order- ly and measured change in sharp contrast to the real world in which technological possibilities are developed and innovations (sometimes) result. Clearly, the Department of Transportation, among other agencies of government, ought to make a special effort to redress the situation in which the Antitrust Division is standing in the way of technological progress that would benefit, on balance, both the private and the cor- porate citizens of the nation. The accompanying tables indicate that antitrust considerations most often appear as disincentives to innovation. This is not surprising, given the nature of antitrust concerns. One of the most unfortunate in- fluences that antitrust considerations impose upon the process of inno- vation is reflected in the often insufficient market aggregation acti- vity found in the transportation field. That is, much of the hardware required by transportation enterprises is produced under conditions of supply in which there are very significant economies of scale in pro- duction, which condition frequently contrasts with the disaggregated nature of the demand for such elements of hardware. Under these cir- cumstances, to make demand sufficiently large at a point in time re- quires some measure of market aggregation--some measure of cooperation between otherwise competing economic units. While there is explicit market aggregation in certain fields of transport at times, especially where public enterprises are concerned and antitrust considerations are nil, for the most part there is far less market aggregation in the trans- port sector than is warranted both by the nature of the supply function for much of what is produced in support of transportation production and by virtue of the benefits of reduced cost that accrue as a result of successful, efficient, and timely innovation in transportation. There can be no doubt that the antitrust "excuse" has been used on many occasions to thwart private sector cooperative or joint activities that would have been benficial both to private entrepreneurs and to the public at large. On the other hand, the antitrust officials of govern- ment, both at the Justice Department and at the Federal Trade Commission, have on occasion been sufficiently sensitive to certain problems that beset specific industries and companies (usually unrelated to innova- tion in any direct way) to relax their otherwise rigid positions. A case in point is the several areas of cooperation between General Motors and American Motors and between Ford and American Motors, obviously designed to keep American Motors from throwing in the towel in the auto- mobile business. The limited history of such flexibility on the part of antitrust officials suggests, however, that an enterprise of industry must be in_ extremis before there is such relaxation of age-old "principles." It is anything but clear that the traditionally great rigidity in anti- trust regulation serves the public interest, and it certainly does not promote the generation of timely and beneficial innovations in many industries, such as transportation. For example, the sharing of test l30

or maintenance facilities by competitors is often thwarted in the trans- port field by antitrust considerations, either imagined or real. In part, the tragedy is that there is so much that is imagined and so little that is real, a situation that ought to be corrected to some considerable extent given the nature of technology in the real world and its contrast with the type of legal system being imposed upon its development and diffusion. Even while the Justice Department is so clearly and properly dedi- cated to the preservation, and even enhancement, of competition in various markets, it is often blind to developments within the transpor- tation field that have the opposite effect. A classic case in point, of course, relates to the railway braking equipment field, discussed previously, in which in the l960s approximately half of the production capability for braking equipment came under the control of a major railroad, which, in turn, sits in the councils of the American Railway Engineering Association and the Association of American Railroads, where the technology that is allowed to be applied to equipment in railway interchange service is determined. This acquisition was permitted without any concern being expressed by the Justice Department or the Federal Trade Commission (or the ICC, for that matter). Of special interest to the antitrust watchdogs should be the grow- ing and unchecked monopoly power of many public enterprises in the United States. But there is no evidence that either the Justice Depart- ment or the Federal Trade Commission recognizes the problem or is con- cerned by it. If they were, they certainly should have recommended some legislation to the Congress by now to enable them to add public enter- prises to their own purview. As was noted earlier, one of the most effective ways to thwart technological innovation is to rely upon design specifications rather than performance specifications when purchasing hardware and software. Surely those concerned with preserving and enhancing competition in the name of antitrust should acquire an understanding of the power of per- formance specifications to assist them in achieving their stated objec- tives. In addition, they can work with other elements of government and with the private sector to see that such power is, in fact, exer- cised to the maximum feasible extent. Certainly, several incentives to innovation would be made more effective even while some disincentives were removed. Once more, DOT can play a triggering and catalyzing role in this regard. It should be understood from the outset, however, that the use of the performance specifications cannot be a one-shot procedure; what is required is a continuing review of performance specifications to make sure that they are revised at appropriate intervals--certainly no less often than every l0 years. In this way, competitive suppliers can be induced to reach the frontiers of technological possibility, which are themselves expanding with time. It is true that reliance on performance specifications, and the award of contracts for hardware and software based upon responses to such performance specifications, will often result in dramatic shifts of patronage from one or serveral enterprises to perhaps only one l3l

enterprise that has distinguished itself in responding to a performance specification. Such a temporary "grant" of monopoly power lies at the heart of the process of innovation and represents one of the most power- ful of all the incentives to innovation. Where concentration temporari- ly results from the application of purchasing techniques that are pro- innovation, the antitrust guardians should recognize this as a reflec- tion of enhanced competition in the long run and not as evidence of the conveyance of permanent monopoly power, which is and ought to be offen- sive to antitrust policy and law. The whole concept of standardization should be viewed in the same general way as the application of performance specifications. That is, physical standards must be set in many instances, but they also should be reviewed periodically to insure that competition is not being thwart- ed and that, indeed, innovation is being promoted. In this connection the recent Hydrolevel antitrust decision is encouraging, especially because of its focus upon the process by which standards are often set. As this case demonstrates, the process can sometimes be subverted and employed explicitly to slow, if not totally discourage, beneficial inno- vation. Market Aggregation Market aggregation has been experienced in many areas of transportation but has not been the subject of the attention it deserves, especially in the context of the process of technological innovation. In part, market aggregation is a difficult concept with which to deal because of the constant fear of antitrust entanglement on the part of private sec- tor entrepreneurs who either attempt to aggregate the market or respond as suppliers to the demands of aggregated markets. Yet, as was previous- ly noted, many items of supply in the transportation field are produced under conditions of great economies of scale, and given the typical fragmentation of the market for transport hardware and software, some aggregation of the various elements of demand is necessary if timely and efficient innovation is to result. The most dramatic instances of market aggregation in the U.S. trans- portation sector have occurred under one of two conditions. First, there are those cases in which the federal government has intervened to become the market-aggregating agent for dispersed enterprises with individual demands that were, to some extent, compatible. An early case in point is the President's Conference Committee (PCC) streetcar experi- ence of the l930s, in which the White House itself was involved in creat- ing a set of more or less uniform specifications, which led to the "mass production" of a large number of PCC streetcars that were distributed throughout the United States and ultimately throughout much of the world. A more recent instance is the Transbus, in which, in the first attempt at procurement, the demands for new urban buses of three major metropolitan areas were combined in a request for quotations in l979. The second set of circumstances leads to market aggregation on a quite routine basis and relates to standardization activities most often l32

associated with the need to maintain interchangeabiiity of equipment. Railroad freight cars are perhaps the best example. Here, the equipment must have a number of common physical attributes if it is to be capable of moving freely throughout the railroad network of North America. Con- sequently, specifications have been developed—many of them 50 or more years ago—that are almost always totally design specifications in character but that nonetheless tend to serve as a market-aggregating agent where freight car components are concerned. Perhaps because this form of market aggregation has been present for so many years, it is allowed to continue with little or no interference either from anti- trust or railroad officials. It is clear, however, that market aggregation in the freight car context has been carried too far in the sense that it has thwarted a number of opportunities for technological innovation that could only have been exploited if there were a departure from established specifi- cations to permit the introduction of new technology even if only on a limited scale initially. (Examples include railway braking and coupling systems.) Obviously, it cannot be held that market aggregation is always catalytic in its effect on the process of innovation. Still, on balance, market aggregation that employs performance specifications that are reviewed and republished periodically must promote innovation, though care still has to be exercised to assure that the result will be net beneficial innovation. As was previously noted, innovation is most often associated with the assumption of risk, both on the part of the sponsors of innovation and on the part of those who accept it. Consequently, one of the most powerful incentives associated with market aggregation is the sharing of the risks of innovation among the various parties to the aggregation process. This cannot be overestimated as a catalyst or incentive to innovation and, indeed, the power of risk sharing is so great that the Department of Transportation would be well advised to catalog all the market aggregation possibilities in the various modes of transportation so as to be able to assess the net cost and net benefits to society that might flow from judicious market aggregation activities sponsored or supported by DOT. Identification and Amelioration of Social Costs Social costs and social benefits have become increasing concerns in the United States as the nation has matured, diversified, and experienced threats to its social and economic development from new quarters, such as the environmentalists. One of the great incentives to innovation in many fields, certainly transportation among them, relates to the improve- ment of what is often referred to as the "way of life" or "life-style." While there is a firm basis for holding that the private sector some- times makes investment decisions and operating decisions based on the amelioration of social costs, in the final analysis it is a function of government to guide the investment decision-making processes in both the private and the public sectors in such a way that social costs are l33

minimized on a net basis. Yet it is clear that the U.S. economy is, by and large, still "flying blind" where social cost and social benefit calculations are concerned. Without adequate and accurate information about social costs and benefits (and the relationships between them), much transport innovation will be wrongheaded, at least where the public interest is concerned. Moreover, several potentially powerful incentives and disincentives to innovate will go unexploited in the absence of an appropriately detailed social cost identification and quantification scheme. Without such "guidance," the innovation propensities for both private sector and public sector executives are being distorted in the transportation field. In part, this is because what may in reality be short-term societal con- cerns can be magnified far beyond what is rational or beneficial, either to the entrepreneur or to the public. There are myriad reasons why government should increase its efforts to identify, quantify, and devise schemes for ameliorating social costs. Improvement of the innovative performance of the transportation sector is but one of these reasons. Standing alone, however, it is sufficient to induce DOT to exert new and strenuous efforts in this direction. Meanwhile, DOT might also consider various means of countering the strong negative effects on certain desirable processes of innovation in transportation that exist because of the growing emphasis on "social costs" and "social benefits," even while society remains largely incapa- ble of identifying and quantifying such costs and benefits with a degree of precision that justifies their use in this way. DOT might well now attempt to devise schemes that, for the present at least, reduce to manageable proportions the disincentive effects of the possible exist- ence of external costs associated with a prospective innovation so that process can move forward on a broader front in the transportation field. DOT can proceed in this regard in a number of ways. One of them relates to the creation of insurance schemes that reduce, if not elimi- nate, losses experienced by suppliers and their customers should a transportation technique be found to be socially undesirable at some point after its adoption and diffusion. While the careful drawing of performance specifications and goals to include externalities-producing criteria is a promising means of reducing the risks of both society and entrepreneurs engaged in transportation innovation, and it is quite clear that this can actually be accomplished without discouraging the "golden geese" innovations, the better course is to promote innovation actively but also to develop mechanisms that permit subsequent "recalls" if an innovation turns out to be socially offensive based on actual experience. l34

DISCUSSANT'S COMMENTS BY HARVEY E. BRAZER The Gellman paper is highly informative and should serve as a useful stimulus to constructive thinking about ways in which public policy can be effective in encouraging innovation in the transportation sector of the U.S. economy. I was especially impressed by the author's demonstra- tion of the folly of government procurement policy that focuses on specification of design details rather than on performance specifica- tions. I should like to have seen more attention devoted to the role of organized labor in influencing the pace of innovative activity. We are all too familiar with the case of "firemen" riding diesel-powered loco- motives and what some would consider "excess" personnel carried in the control compartments of commercial aircraft. But what is much less obvious is the appropriate set of incentives that would serve to remove such barriers to innovation. Much the same may be said of management incentives with respect to innovative activities—and Gellman does deal with this problem more fully—and the relative rewards of risk avoidance as opposed to risk taking. Perhaps innovation is strongly discouraged because rewards tend to focus on outcomes to the exclusion of inventive- ness in attempts to improve outcomes, irrespective of results. In other words, we may be strongly in need of a means of providing a payoff for innovation as such, in a way that would drastically reduce the costs, and therefore the risk, of failure. In discussion of innovation in the United States, it should be recognized that, through outlays on research and development that currently run at about 2.5 percent of GNP and by other means, the post- World War II years have witnessed an enormous outpouring of new products, new production and distribution techniques, and so forth. The problem confronting the committee stems largely from the fact that innovation in the transportation industries has not kept pace with that in other sectors. Clearly, innovation need not involve capital outlay requirements, but wherever new technology is the means to innovation and that tech- nology is embodied in capital assets, capital outlays associated with innovations may be enormous. U.S. government policy in the field of taxation has been cognizant of this for at least 25 years, and several major steps have been taken to improve the profitability of investment in depreciable assets other than structures and to increase business cash flow. In l954 the Congress permitted the use of such accelerated methods of depreciation as double declining balance and sum of the year's digits; in l962 the investment tax credit was introduced, at a rate of 7 percent, increased to l0 percent in l976; and new shorter l35

lives for depreciation were introduced in l962 and further liberalized in l97l with the adoption of the "asset depreciation range" approach to depreciation. Taken together, these changes in the tax treatment of business capital assets now provide some $50 billion a year in addi- tional cash flow to American enterprises, approximately equivalent to a drop of 40 percent in the federal corporate income tax. And this is exclusive of recent reductions in the tax rate itself. There is now much discussion, in Congress, in the administration, and in business and academic circles of the need for further tax help for the supply of investable funds. A strong candidate for enactment is the "3-5-l0" bill. It would permit vehicles to be depreciated in three years, machinery and equipment in five, and structures in l0 years. That such a measure would encourage and facilitate innovation is undoubtedly true. But what is not clear is whether or not it would be efficient public policy. With respect to transportation, the current picture for much of that sector is such as to suggest that tax concessions are not the answer. Few railroads now pay appreciable amounts of federal income tax, so that additional depreciation allowances, if made available and if taken, would serve only to increase accounting losses. Two of the big three auto firms will report huge losses for domestic operations in l979, and the outlook for all three for l980 is bleak. Thus, depreciation that is too liberalized is certainly not likely to be helpful here in the near-term future. On the other hand, other segments of the transportation sector continue to be profitable, including air- lines, trucking, pipelines, and their suppliers of equipment and other capital goods. And for these, of course, liberalized depreciation would be helpful. More important, in general, however, is the question of whether the tax route is the most appropriate, most effective means of achieving the goal of stimulating innovative activity in transportation. For the reason just suggested it is quite obviously not useful to some of the major parts of this sector, and it may not be the most desirable approach for any part of it. This follows, in part, from the fact that accelera- ted depreciation, the investment credit, and across-the-board tax cuts are not and cannot readily be designed to be directly related to innova- tion. That is to say that the same tax break is afforded to the firm that acquires more of the same old machinery to produce, using the same methods, the same old products, as is afforded to the firm that acquires machinery that embodies a revolutionary new technological breakthrough. Thus one may expect that for each dollar of tax concession that facili- tates innovation there may be several that do not. Apart even from innovation, it is not clear that tax policy of the kind now widely advocated is very effective in stimulating investment. A number of surveys and econometric studies conducted in recent years suggest that each dollar of reveue foregone through the investment credit and acce- lerated depreciation tends to be associated with anywhere from some- thing less than a dollar to a little more than a dollar of new invest- ment. At this rate the Treasury would seem to be getting a rather poor bargain—certainly if the revenue cost were to be compared with the volume of new investment representing innovative change. l36

It seems to me that we have gone to the same old well to often with too little to show for our efforts. Surely the fact that virtually no one concedes that either our investment or our innovation needs have been met through tax policy measures of the past must suggest that some innovative thinking is needed at least as much in this area as in any other. The answer may be found in the form of grants tailored to pro- vide the needed reward system for innovation of all kinds, whether or not they involve capital formation. But before new public funds are committed to this endeavor, in any form, we should be reasonably sure that we have made a strong, intelligent effort to remove the immeasura- ble institutional barriers to innovation that Gellman and others have pointed up so well. l37

PROCUREMENT AND INDEPENDENT RESEARCH AND DRVFLOPMF.NT REMARKS BY ALLEN E. PUCKETT Our panel is concerned with independent research and development (IR§D) and procurement. I take that to mean government procurement. I am not sure about all the connections between that topic and the field of trans- portation, but we will explore that in our panel meeting. IR§D generally, in the industrial world, means that element of research and development that is conducted entirely at the initiative of the company. It may be conducted without respect to any particular contract support or orders from the government, from other companies, or from other customers, and it is a research and development component determined by management to provide for the future of the company. In plain English, the purpose of independent research and development is to improve old products, or to develop new products, and services, as the case may be. It is not directed or controlled by customers at all, in the direct sense. Nevertheless, properly managed IR§D has a primary purpose, and that is to be responsive to customers: to be responsive to perceived customer needs and, in many cases, to anticipate customer needs or perhaps even in the most important cases to create new customer needs. I do not need to go through the long list of things in that cate- gory that you all know as well as I do. Some of the most important real innovations in the country came about before a customer need existed. In fact, an important part of the innovation process was to create, eventually, that demand in the customer world. So that is the IR§D role. How is it paid for? In the end, it has to be paid for by our customers, if one speaks from the industrial point of view. In other words, this follows the old first law of economics: there is no such thing as a free lunch. Sooner or later, one way or another, the price of all the IR§D that we do must appear in the price tag of the customer. But now, as a practical matter, the mechanics of how this is done is a matter of accounting procedure. I will let you in on another secret that may be known to many of you. Accounting is not an exact science. When I was a much younger engineer, I suppose I had some illusion that accounting was a matter of arithmetic, that the rules were quite clear, and that everyone, when they had read the book, knew how to do their accounting. It turns out that that is not true. In fact, we have a little story at our place about a very short form method of interviewing applicants for various categories of jobs. l39

Three very important professions, of course, are engineering, law and accounting. When the engineering applicant shows up, we ask him, "What is the sum of one plus one?" If he says "two," he is a pretty good engineer, and we hire him. When a lawyer shows up, we ask him, "What is the sum of one plus one?" If he says, "Well, it all depends," he is a pretty good lawyer, and we hire him. Then when the accountant shows up, we ask him, "What is one plus one?" and if he says, "What would you like it to be?" then he is a good accountant. So the many techniques by which the costs of IR§D are handled in the accounting procedure are various and wonderful. But essentially, IR§D in any corporation eventually shows up in an account called overhead. Overhead is sometimes regarded by governmental authorities as a kind of pejorative term. There is the view that overhead is something that we should not have and further, that a measure of your management capability is the size of your overhead, and the lower the overhead, the better you are. As a side comment, I have to point out that that is not true. My idea of the perfect manufacturing plant is one with about l0,000 percent overhead, because there could be one man in there on direct labor and everything else could be automated, and that is overhead. But that is a different topic. But the point is that sooner or later all IR§D charges go into overhead. The cost, the overhead account, in some way is allocated against cost of sales. That is the way the customer pays, eventually. Put another way, IR§D in any segment of American industry is a normal cost of doing business. It is an essential part of maintaining the health of the company and of providing for the future of the company. It is included as a necessary element in the cost of sales. It is in the price of the product. However, the price of the product may not necessarily be directly related to its cost. That is another matter. But in any event, if we do our accounting properly, the element of IR§D is included in the cost of sales. I will now switch to the special area of government procurement. Currently, we have some extra rules. In the case we are considering, in these complex relationships between industry and government in the government procurement process, for good reasons the government negoti- ates with us generally on our overhead rates. That is not necessarily true if they are buying shoes, buying a catalog item, or buying many commercially available items. However, in the more complex endeavors that we are talking about here, the government does negotiate overhead rates. In fact, they find it desirable to negotiate, in particular, this little segment known as IR§D. One of the topics, then, of great interest to all of us who are involved in government procurement concerns the rules, the policies, and the practices that accompany the negotiation of that segment of our IR§D that may be legally, legitimately included in this overhead rate. A segment that represents part of the cost of sales, which in turn we are going to allocate to the price of our product. The end result is that the IR§D dollars a company spends as a normal cost of doing business may not be allowed completely in the price of its l40

products to the government. The government is still going to pay in one way or other, but it simply does not come out as a fixed, finite part of the allowable cost. Some interesting questions of policy are involved in this matter of government treatment of IR§D. It has been a controversial topic for many years, and I really do not want to try to give the conclusions of the panel because I do not know what they are going to be. I do think, though, that this is an appropriate time to bring together some experts in this field, on both the government and the industry side, and to review what it is we have been doing and whether we think it still makes sense. To return to the relationship of all this to transportation, I said at the outset that I am not sure whether there really is a connec- tion. I am not an expert in transportation at all, so I come here with no preconceived ideas. I did think, though, that I should get some feeling, some perspective, on this vast area that we are talking about— the national transportation picture. The things I learned in scanning the l977 DOT report on Trends and Choices in transportation—a fascina- ting document—are probably well known to you. Some of them were certainly a bit surprising to me. One of the speakers mentioned that the transportation industry, or enterprise, the services, the manufacturing, and the public users, takes up about 20 percent of our gross national product. That is a remarkably large number, around $600 billion in current terms. I learned that about half of the GNP fraction is related to passenger transportation, and about half of it to freight. That was interesting. The next thing I learned, and this may be very well known to all of you but it was a little bit of a surprise to me, was that in the passenger area, over 90 percent of all passenger miles are provided by private automobiles! It may be over 90 percent depending on how one does the calculations. So the remaining l0 percent or less of passenger miles is supplied by airplanes, buses, railroads, and boats, I suppose. To the extent that we are interested in passenger transportation, the inescapble fact is that the automobile totally dominates the scene. We could double urban mass transportation patronage, and it would hardly affect the nature of automobile transportation at all. At least, that is the pre- liminary feeling I get. Transportation is a big business. There is a great amount of leverage in automotive transportation; an enormous industry is involved. I then took a look at the Department of Transportation to see what our leverage and the interconnection are there. The Department of Trans- portation's budget is about $l7 billion a year. Of that, something around $l2 billion is dispensed in the form of grants. A large part of that is in highways, and another part of it is in urban mass transpor- tation. So, in that sense the DOT is really just a pipeline, a pass- through arrangement, for this money that flows out in some fashion, but apparently is not involved at that level, in government procurement. Finally, I saw the item for R§D. Now, there may be some other elements of R§D hidden in the budget that I could not find in a short l4l

study, but I will give the item that I noticed, in billions because we started with billions. We started with $600 billion for the whole of transportation. I got down to $l7 billion for the Department of Trans= portation. The R§D segment is about $0.38 billion. That is not a very big number. Just by comparison, the DOD budget component in R§D is about $l2 billion this year, or something on that order. That is out of a total budget of about $l20 billion. It may not be fair to draw those compari- sons, but I will draw them anyway. Once again, that raises in my mind the real questions regarding the role of IR§D, of government procurements, and of government R§D generally with respect to transportation. From the few things I have said, it would be very tempting to draw the conclusion that the rela- tionship is somewhere between small and zero, but that is probably not right. I want to take a more positive approach to encourage the panel. The one thing that is overwhelmingly true about IR§D anywhere, in any industry in the United States, is the enormous leverage that it has on the future of the company, of the country, of the economy, and of the state of the industry. The leverage is tremendous, and therefore, even attention to small numbers and faint connections may have real value. It may be that innovation, and, of course, here I am thinking particularly of the technical aspect of it, may depend much more on factors in the private sector, but the stimulus of the government's interest in IR§D—interest or the lack of it, as the case may be—does in turn have an effect on industry, generally. That is particularly true in the high-technology industries, where very often the most difficult problems that we attack, the impossible jobs that we attempt to do, really provide the stimulus or the pressure to create. I am distinguishing here from innovation. I mean the stimulus to create, to invent, and to conceive the new solutions that open the door to many applications that were not initially in mind. So the leverage is there, and the connection is there, and I think we have an interesting opportunity to explore it in this panel. A VIEW OF U.S. GOVERNMENT CONTRACTING POLICIES AS THEY RELATE TO THE SUPPORT OF INNOVATION BY WILLIAM L. RODENBAUGH AND W.B. GIST INTRODUCTION The dictionary tells us that an innovation is the making of a change in something established. The change is usually understood to be the bring- ing in of new ideas, methods, or devices. Innovations in commercial l42

and industrial establishments are mostly analogous to the biological process of evolutionary change through mutation, though less frequent revolutionary changes do occur. Market pressures are normally adequate to motivate, develop, and insure the survival of the fittest innovations provided that they are intially seen as sufficiently advantageous to outweigh the risks and inconveniences of change by both the supplier and the user. Our concern here is with those exceptional situations in the transportation industries in which these market pressures are inadequate or inappropriate to provide innovative improvements at a pace perceived to be required by the public interest. The National Transportation System is a classically mature esta- blishment with formidable "risks and incoveniences" to both the supplier and the user of any change. Consider the railroads. There is little potential for market growth or penetration by outsiders. Well-developed and inflexible systems are established with an entrenched industrial base. Restrictive regulations and equipment standards abound. The labor force has the demonstrated power to negate the benefits of labor- saving innovations. Regulated tariffs and costs not entirely under the control of management have kept profits too low to support an adequate R§D base. The cost and risk of any change in this system discourage the adoption of any but the most modest evolutionary innovations and provides little incentive for innovative effort on the part of equip- ment suppliers or users. The level of maturity and stagnation attained by the nation's railroads is unique, but most of the industries affected by the Depart- ment of Transportation have some of the innovation-stifling characteris- tics of the railroads. The air transportation system is at the other end of the spectrum among revenue carriers, but even here, the signs of encroaching maturity are evident. The government has several vital roles to play if innovation in these systems is to be encouraged: l. Providing directions or goals for innovators based on a long- range overview of societal/economic needs. 2. Augmenting the commercial incentives for innovative effort (and eliminating disincentives). 3. Overcoming or eliminating regulatory and institutional road- blocks. 4. Aiding in the capitalization of innovative experiments and systems. Inventiveness and genius for innovation can be discovered almost any- where; it is assumed that the likelihood that it will be discovered is proportional to the number of potential sources attracted and enlisted in the effort to solve any particular problem. The likelihood that any given innovation will be able to fulfill the purpose for which it is intended is probably enhanced if the innovator has appropriate skills and knowledge and facilities to apply to the problem solution. Procurement policies are conceived not only for the purpose of establishing orderliness and consistency in the administrative procedures for managing the purchase of commodities from vendors with prudence and wisdom (as is suggested by the dictionary's definition of the word policy). They also serve as instruments for accomplishing social, l43

environmental, and economic goals of the Congress and the administration that are not necessarily related to the quality of the end product. Inevitably, they are a factor influencing the responsiveness of commer- cial firms and universities, which could be leading sources of innova- tions sought by the government for the purposes of improving national transportation systems. This paper will reflect on the existing procurement policies of the government and how they might be viewed by potential suppliers of inno- vative concepts, studies, developments, and products as providing either incentives or disincentives for applying their talents and resources to government-contracted programs. HOW FEDERAL PROCUREMENT POLICIES ARE VIEWED BY POTENTIAL INNOVATIVE CONTRACTORS One of the ways in which the government can give directions to and pro- vide incentives for innovative efforts is through improvement of federal procurement policies and practices, especially in regard to research and development efforts undertaken by industry in support of government objectives. These policies are continually being changed to meet newly perceived needs of the government as a customer and as a guardian of the rights and a dispenser of the benefits for every segment of the population and of the economy. It sometimes happens that the policies adopted turn out to be counterproductive with respect to obtain- ing the best that industry could provide in the way of innovative contri- butions. The very immensity of federal expenditures and the correspond- ingly heavy stewardship responsibilities of both the executive and the legislative bodies dictate that procedures designed to protect the public treasury be extremely complex with provisions that are sometimes onerous to contractors. Some of the disincentives in government procurement policies stem from the government's laudable objective of maintaining multiple compe- titive sources for a particular service or product. When this objective is sought by applying such devices as a conflict of interest clause to minimize the competitive advantage gained by a particular contractor from work performed on a prior government study contract, the result tends to discourage early competitive participation and thus inhibit innovative effort. Other problems stem from the government's legitimate desire to get full value from government contracts: services and products that meet the government's quality expectations and currently perceived needs (which frequently change during the performance of long-term government contracts), with on-time delivery and no cost overrun. These objectives have led to intrusion into the contractor's customary managerial func- tions and prerogatives by government inspectors, auditors, and reviewers. This adds significantly to administrative and clerical costs incurred by a firm and limits the ability of company management to control speci- fications and costs within the originally contracted values. 1 Council of Defense and Space Industry Associations (CODSIA), "Study Costs Unique to Government Contracting," December 8, l97l. l44

These practices, plus the need to minimize procurement costs when effective competition is lacking or when a unique product or service is single-sourced, cause most government procurements to be negotiated with a cost-based price. This invokes a fundamental issue of how much mar- gin to allow for profit and reinvestment by the company. Company re- investments of margin into company-selected and company-managed research and development for the purpose of maintaining or enhancing the future competitiveness of the company are thus subjected to separate identifi- cation and audit as cost elements by the government. This element of cost has been labeled independent research and development (IR§D) by some government agencies, and its allowance has become a political issue. The remainder of this paper will enlarge upon the relationship of these procurement policies to the interests of potential contractors in government-sponsored programs, leading to recommendations that would stimulate participation by university and industry innovators. Setting Goals In providing a direction for innovative research and development, the government must be careful to identify the problem precisely (and not attempt to specify the solution) in order to obtain most effective ap- plication of the independent inventive genius of industry and academia. Good problem solutions are not always apparent without some exploratory research and evaluation. Contracted studies to aid in the process of winnowing the good ideas from the rest should be obtained from qualified researchers with applicable knowledge, skills, and technology. The Department of Energy has recently imposed a policy that excludes the best-qualified firms from participation in technical evaluation and consulting services on the basis of a conflict of interests. This organizational conflict of interest policy is being considered by other government agencies. Most firms in industry engage in R§D activi- ties only as a means of obtaining a competitive advantage in their future products or technical capabilities. This conflict of interest policy can be a strong deterrent to their participation in a process where they have much to offer and where the most significant decisions on potential innovative opportunities are frequently made. The govern- ment believes that participants in early studies leading to a product development or procurement will obtain a competitive advantage over nonparticipants. Why else would a firm of innovators risk the dis- closure of its good ideas to potential competitors during the course of such preliminary evaluations? The government also seeks to avoid biased assessments of alterna- tive approaches by this proposed policy, but biases can be discerned and allowed far more readily than incompetence and lack of realism. Parallel competitive studies by firms with a vital interest in the results may be a better means of providing balance. l45

Providing Incentives (and Eliminating Disincentives) for Innovation Participation It is in augmenting the commercial incentives for innovative effort that the government's procurement policies have the most impact. DOD has successfully encouraged dramatic innovations in military systems through farsighted policies on R§D procurement, including directly funded in- dustrial R§D, and NASA has had similar success with similar policies in both space and aeronautics. Commercial spin-offs from the DOD and the NASA-sponsored innovations have provided benefits to the nation far beyond the original objectives of these agenices and far beyond the level that would have resulted from commercial incentives alone. The complexity of the federal procurement policies is such that the sheer volume and the administrative costs associated with keeping abreast will deter some commercial entrepreneurs from becoming involved with government business. Some of the major differences between commercial business and business under government contracts are as follows: l. In the government marketplace, there are no permanent commit- ments—the government order that a business receives is incrementally funded and, if performance is to be carried out over several years, is subject to annual congressional program review as well as continuous reviews by several layers of "yes/no" decision makers who can decide to scrap a program or alter its direction after it has started. The company new to government business will find such potential program vagaries accommodated in its contract with special terms and conditions, e.g., termination for convenience, limitation of government obligation, and changes. In the commercial marketplace, the usual pattern is for the seller to carry out performance and delivery as ordered by the customer with a reasonable certainty of contract continuity as agreed by the parties at contract signing. 2. After selection of a supplier through examination of his pro- duct and his specifications for a potential product, the commercial purchaser places an order and expects the supplier to manage his own business to assure contract performance. Except for specific and agreed-to contract provisions, the Uniform Commercial Code governs the transaction or its outcome in the event of dispute. The government purchaser instead will impose his own standards for such fundamental business activities as quality control, production control, and sub- contract purchase routing, in "courts" (Board of Contract Appeals) using federal law/regulations--Armed Services Procurement Regulations, Federal Procurement Regulations, NASA Procurement Regulations (ASPR/FPR/NASAPR). If the government contract is negotiated, the supplier is also subject to the rigid cost control/financial system disciplines encompassed in the cost accounting standards and the cost (dis)allowance criteria in section l5 of the ASPR. To compete for a nego- tiated contract with the government, the potential supplier must furnish cost projections in extensive detail and provide government auditors with access to the supporting documentation for all direct l46

and indirect cost.2 After completion of the financial audit and several surveys of the business, e.g., purchasing, production, and tech- nical, the give and take of the negotiation process is initiated, in which the government's objective seems to be to obtain the lowest possi- ble (not probable) cost to which a statutory or regulatory (low) profit rate is applied. These regulations are imposed as well on the winner of the contract and affect each subsequent change negotiation. At frequent intervals during performance, business reviewers are sent to the supplier's plant, or for larger, longer-term contracts, they will be resident in the place of business of the supplier as a permanent staff of auditors, inspectors, and reviewers. 3. In addition to the aforementioned imposition of government controls on the "business" aspects of the enterprise, the would-be government supplier must be aware of the difference he may expect from commercial practices due to the impact of the technical and product acceptance criteria used in most government procurements. Commercial customers expect to receive and will inspect and accept a product in accordance with agreed-upon practical specifications-performance criteria. Express warrantees define the supplier's after-delivery responsibilities. Because the government accepts greater after-delivery responsibility for the product, the supplier to the government usually finds that his contract not only defines what he is to deliver and how it will be inspected and accepted, but also will control many of his technical and manufacturing activities: the sources for certain mater- ials and component parts, the in-process inspection techniques, the ability to change the physical configuration of internal parts, the drafting practices, and more. If disagreements arise, the supplier's recourse is first through one of the government's administrative Boards of Contract Appeals and not through the courts suppliers use in most commercial disputes. 4. Another difference between commercial and government business is in the area of rights in technical data. Commercial purchasers of goods and services rarely expect any form of data rights as a condition of purchase; in the special cases in which such rights are sought com- mercially, license arrangements providing for special compensation are the usual approach. Contrary to the commercial approach, the govern- ment usually seeks to obtain the rights to make or have made for its own use any product (or its parts needed for repairs) developed or modified under a government contract. Before entering into a contract, which may jeopardize any patent protection, know-how, or unique process the business may have developed on its own prior to a government con- tract, the prudent commercial businessman will obtain counsel from a competent legal authority specializing in government contracting as to those precontract technical rights that may flow to the government as a result of the contract. *• CODSIA identified l7 different types of audits and reviews for which there are no comparable commercial work costs or which exceed similar types of costs on commercial business in their "Study of Costs Unique to Government Contracting," December 8, l97l. l47

On the other hand, government contracts have a number of advantages over normal commercial contracts that make them very attractive. Govern- ment contracts may provide for special financing arrangements, i.e., progress payments, without cost or price concessions from the supplier. This is especially helpful when a research and development program requires many years or imposes large costs. Also, while government contractors may be concerned about abrupt contract terminations, they need not be concerned about the purchaser's bankruptcy. Most important- ly, the results of technical efforts supported by government development funds may be introduced into the commercial products of the business to enhance their competitive value or lower their costs. The additional volume of a government contract may also improve the utilization of fixed investment through lowering the costs on a continuing commercial business. Perhaps the most widely debated issue of federal procurement policy is the propriety of compensating contractors for the independent research and development that they conduct in order to retain or enhance their future competitive capability. In dynamic and competitively innovative industries, research and development is essential to continued survival. In these industries a company must consider reinvestment of margin into research leading to the development of future products or new markets as part of the cost of doing business. Margins must be maintained high enough to provide for this reinvestment, but at the same time prices must remain competitive on current products. Thus the managements of these firms are forced to use great care in deciding how much effort to apply and in selecting research activities with a potential payoff. Such decisions are "the most difficult, and in the long term, most significant decisions of management in any enterprise, because in no other way does a company put its future on the line to the degree that it does in making such determinations. "•* This internal discipline for assuring reasonableness and appro- priateness of R§D based on long-term competitiveness of products as well as short-term competitiveness of prices works very well under most cir- cumstances. But when effective competition is lacking or when a sole-source procurement is necessary owing to uniqueness of a product or service, prices of government purchases are negotiated on the basis of actual cost plus a "reasonable" profit. In negotiated government procurements the amount of profit must not exceed specified statutory limits, and contract negotiators usually settle for profits significantly below the statutory limits and below commercial profit levels achieved after re- investment in independent research and development. Thus, if allowed at all, the IR§D reinvestment must be treated as a negotiable and auditable cost along with all other costs, and as such the government must be satisfied as to the reasonableness of the level of this cost and the appropriateness or relevance of the activity in relation to the •* AIAA, "Recommendations to the Domestic Policy Review of Industrial Innovation," AIA, February 2, l979. l48

government's interests. (Independent Research and Development, or IR§D, is a term devised by the Department of Defense and used by federal agencies to identify a contractor's basic and applied research, develop- ment, and concept formulation studies performed under circumstances other than a government-sponsored arrangement such as a contract or grant.) The Department of Defense has evolved an elaborate procedure for controlling IR§D costs recoverable on their contracts. The procedure involves prenegotiation of allowable dollar ceilings (if the contrac- tor's prior-year "payments" for IR§D plus bid and proposal (B§P) exceed- ed $2 million5 on both IR§D and B§P. In establishing the ceiling the government takes into consideration a technical evaluation of the pro- posed IR§D projects and the potential military relationship of the projects. The ceiling almost always is below the actual IR§D and B§P expenditures of the company. Subject to progress reviews and reports and audits of all projects included under the ceiling, the government allows only a share proportional to the ratio of DOD sales to the total sales of the company to be recovered in the price of products sold to DOD during the year covered by the agreement. These policies on IR§D reimbursement have apparently been an out- standing bargain for the DOD. In l975, Director of Defense Research and Engineering Malcolm R. Currie stated in congressional testimony that, "In l974, on the average, 92 percent of all IR§D projects were directly relevant to DOD interests while, on the average, DOD paid for only 39 percent of the IR§D effort incurred. For this discounted pay- ment, the government is able to maintain the most advanced technology and innovative systems in the world." Industry regrets that the government has elected to put many restraints on IR§D. The Aerospace Industries Association of America (AIAA) has stated to the Domestic Policy Review of Industrial Innova- tion that: It is to government's advantage to preserve the independent nature of a contractor's research and development effort. Independence permits a firm to apply its resources to those technologies and programs in which its capabilities are high- est and which, therefore, will provide greatest benefits to both firm and customer. Government control that inhibits the flexibility of industry to respond to the changing market environment is clearly an adverse influence. It makes government partly responsible for the success or failure of industry and pre- sumes that government possesses some sort of omniscience that has never been demonstrated. 4 Defense Procurement Circular (DPC) 90, effective, January l, l972. 5 For companies not required to negotiate advance agreements, allowable IR§D and B§P costs are determined by using a formula based on previous years' costs and sales. l49

The present governmental method of recognizing IR§D costs may restrict the amount of IR§D costs which can be recovered under a company's government contracts. Under Public Law 9l-44l, only those projects considered by DOD to have potential military relationship are considered for cost recovery—and then only within a ceiling established by negotiation or formula, depending upon the company's previ- ous recovery of IR§D costs. While industry has learned how to comply with these restrictions, there are serious drawbacks in the present method. The basic concern is that the method really does not recognize IR§D as a legitimate cost of doing business; it implies that IR§D is dispensable when it is not. The price of every company's products should properly include the company's proper cost of doing business and each customer should pay its fair share of that cost. As a customer, the government is neither buying IR§D as a commodity, nor is it supporting or subsidizing IR§D; instead, it is buying goods or services, the prices of which should contain a proportionate allocable share of all indirect costs. Legislation under which the government enjoys a preferred position, free of the obligation to pay a pro-rata share, un- fairly discriminates against other customers. It is particular- ly unfair when the government is in a position to influence legislation to its own advantage. To the extent that government refuses to recognize such costs in its prices, government receives an un- warranted discount on its purchase. To the extent that government does not pay, the burden is shifted to the stockholder in terms of reduced return on investments; or where the company also produces for the commercial market, the extra burden may result in higher prices for the con- suming public, thereby weakening the company's competi- tive position in the market. The company is thus persuad- ed, often even forced, to shift away from government business, because the return on investment is not compara- ble with the return from other markets. IR§D controls, similar to those used by DOD, are employed by NASA, DOE, EPA, and some DOT agencies, but some other agencies allow no recovery of IRSD expenses. At the very least, the government should standardize IR§D recovery regulation for all agencies and allow govern- ment-wide relevancy tests. For instance, IR§D relevant to DOT interests should be allowed under ceilings established by DOD and all other govern- ment agencies, and it should be recoverable on all government contracts performed by the company. CONCLUSION AND SUMMARY OF RECOMMENDATIONS The government has important roles to play in fostering innovation in the nation's transportation systems due to the limited incentives and l50

disproportionate risks that stifle commercially motivated innovative efforts, particularly in the federally regulated transportation systems. In providing guidance and direction for innovators and in augmenting commercial incentives to attract innovative entrepreneurs, the govern- ment's procurement policies can be either a positive or a negative fac- tor. It is in the government's interest to obtain the active applica- tion of the country's best industrial and academic knowledge, skills, and ideas to the nation's perceived transportation problems. Toward this end, implementation of the following recommendations could make a significant contribution. They are operable within the confines of the need to use cost-based price contracting. RECOMMENDATIONS The government should do as follows: l. Provide directions or goals for innovators based on a long- range overview of societal and economic needs. 2. Rely on parallel exploratory and feasibility evaluations and assessments in establishing innovative goals and potential problem solu- tions. The organizations contracted for this effort should include firms with applicable skills and knowledge in the field. These firms should not be prohibited from any follow-on development or production opportunities. 3. Strive to simplify and standardize the procurement regulations used by various government agencies with the goal of minimizing non- productive management/administrative burdens upon the contractor and unwarranted cost to the government. 4. Fully compensate contractors for all overhead expenses incurred due to government-imposed contract management requirements. 5. Fully compensate contractors for intellectual property acquired by the government, and only those properties necessary to protect the legitimate needs of the government should be acquired. 6. Recognize that IR§D is a legitimate cost of doing business and that its cost should be recovered in the prices of the company's goods and services. 7. Provide that IR§D be truly independent as regards the performer's choice and execution. 0 Recommendations 6 and 7 are selected from those made by the AIM to the Domestic Policy Review of Industrial Innovation. l5l

DISCUSSANT'S COMMENTSl BY JAMES E. CARPENTER I can offer no specific criticism of the paper. I certainly agree that such issues as conflict of interest, the recognition of differences between commercial and government business, rights to technical data and intellectual property, and the treatment of Independent Research and Development/Budgets and Programs (IR§D/B^P) are significant and should be addressed by the panel. Particular emphasis should be given to IR§D; if any procurement issue needs clarification and needs to be placed in the proper perspective, that is the one. Such indirect costs must be recognized as a legitimate cost of doing business, the existing DOD relevancy requirement needs reexamination, and the critics should recog- nize that such activities have averaged only 3 l/2 to 4 percent of costs over the past several years. Several elements of overhead are much larger and, in many cases, less productive. Some criticism of IR§D/B^P might be valid and merit discussion by the panel. First, there is some concern that the present procedures discriminate against small firms in particular and any firm attempting to enter the government marketplace for the first time. Second, some data indicate that R§D activities con- centrate on the downstream, heavy development end of the spectrum rather than on the early-research, concept formulation phases. Is this good or bad? One final point regarding the paper: The suggested recommenda- tions sound very much like OMB Circular A-l09, and this should be acknowledged. It should also be made clear that the Office of Federal Procurement Policy (OFPP) in OMB (responsible for the development and implementation of A-l09) has ongoing activities directed toward the resolution of many of the issues highlighted in the paper. The present status of these OFPP initiatives should be made available to the panel. The panel's charter is to examine and make recommendations on issues that relate to the package of and interrelationships among procurement, IR§D, innovation, and transportation. To place these four elements in a context suitable for a meaningful discussion, I would suggest the following approach. The federal government support of R§D and innovation usually falls within one of three main categories of rationale: (l) support of the nation's technology base (basic research, special facilities, etc.); (2) the development of new products or services to be used by the sup- porting agency (DOD, NASA); or (3) the development of new products or services for public consumption (DOE, DOT, etc.). By the inclusion of The views expressed in this discussion paper are those of the author and do not necessarily reflect the official position of the National Science Foundation. l52

transportation as a specific in the panel's charter, our focus should be on item (3), which is the issue of commercialization to meet civil agency missions. If this premise is accepted, the following agenda items seem appro- priate for panel discussion, analysis, and resolution: l. The general problem of commercialization is a very recent and growing issue in the science and technology policy area. Some studies have been completed, and many are now in progress. It is becoming apparent that the innovation process supported by civil agencies is much different from that successfully experienced by those agencies concerned only with obtaining products and services for their own use. One of the reasons for this difficulty is cited in Larry Goldmuntz's paper prepared for another workshop panel. The user is not the buyer, the buyer doesn't pay for it, the payer usually doesn't buy it or use it, the operator is a professional who doesn't use it, pay for it, or buy it. A second problem area was addressed by the Charpie Task Force Report to the Department of Energy, February l978. If DOE's objective is commercialization, it should be heavily staffed with entrepreneurs rather than technocrats, R§D managers and their economic advisors. An analysis of DOE's roster of several hundred R§D executives revealed that only eleven had commercialization experience. As an example, the Task Force observed that most DOE contracts were overmanaged, and were therefore much more expensive to the public than they need have been. The panel should be cognizant of the fact that many of the experi- ences learned in the DOD/NASA-type marketplace may not be useful for transportation commercialization. Notwithstanding the above, there may be elements of procurement practices, if properly applied, that could be helpful. Included might be OMB Circular A-l09, treatment of IR§D/ B§P, patents, background data rights, unsolicited proposals, and others. As long as the marketplace differences are recognized, the panel could conceivably develop worthwhile recommendations in the commercialization area. 2. The implementation of "The Federal Grant and Cooperative Agree- ment Act of l977" (PL 95-224) could have a mjor impact on commercializa- tion by civil agencies. A major mechanism for dispensing DOT funds is by grants to state and local governments. They, in turn, contract with the private sector for goods and services. A second mode is by con- tracting directly with the private sector, usually for demonstration- type projects. According to PL 95-224, some changes may be required in the mechanism for fund transfer. For example, if the cooperative agree- ment is utilized for demonstration programs, there may be changes in such procurement-related issues as patents, IR§D, cost sharing, profits, background data rights, etc. The panel should recognize the difference between acquisition and assistance and the impact of the assistance mode on innovation in transportation. 3. Much of the direct commercialization activities of the civil agencies has been via demonstration programs. Overall, the results have l53

been poor. The DOT experience should be reviewed by the panel to deter- mine reasons for failure and to assess possibly the merits of the demonstration approach and perhaps changes in procurement procedures that might result in improving the chances of program success. 4. The Department of Commerce has recently completed a Domestic Policy Review (DPR) on industrial innovation. The final report of this study is now being reviewed at the highest levels within the ad- ministration. It is not known at this time which of the DPR recommenda- tions, if any, will be supported by the president for implementation. However, a panel review of the industry inputs to the study may be worth- while, since many of these inputs relate to procurement, IR§D, and inno- vation. 5. Several of the papers submitted to the workshop panels high- lighted the possibility that a major cause of DOT difficulties may be organization. For example, and again I quote from Larry Goldmuntz's paper: There is no political stability in DOT. There have been five secretaries of transportation in l0 years, as well as five UMTA administrators. Not one of these rose through the ranks, a sign of a badly managed enterprise. Each secretary tended to reverse the policies of his predecessor. Volpe wanted to mandate air bags, Brinegar felt the decision should be left to the private sector, Coleman wanted to test air bags, Adams mandated air bags, and now we have Goldschmidt. Volpe wanted to mandate Transbus, Brinegar wanted to leave it to the transit operators, Coleman selected Transbus para- meters acceptable to operators and manufacturers, Adams man- dated more stringent parameters, and operators and manufac- turers "got off the bus." Another organization concern, probably more appropriate for panel consideration, is the fact that there is no single office at the assistant secretary level, or above, responsible for the department's technology program. This may or may not be a problem, but the existing organizational structure of the department should be reviewed by the panel to determine if organizational changes might be helpful in impro- ving the DOT procurement process as it related to technological innova- tion. The above five agenda items are certainly not all-inclusive and do contain some coupling between items. Hopefully, if considered by the panel, they can serve as a springboard for meaningful panel discussion resulting in recommendations to improve the DOT technological innovation process. l54

TECHNOLOGY AND RgD PQI.TrTF.S TO STIMULATE INNOVATION REMARKS BY HERBERT D. BENINGTON The scope of our panel includes consideration of the overall federal research and development program in transportation and also the federal policies that can affect research and development done in transporta- tion elsewhere. It is not a part of our job to look at the individual industrial R§D programs and to judge what they should do in the way of technology and R§D. I think I realized what the general tone about federal R§D was going to be when I noticed that this is the first time in a meeting such as this on innovation that federal R§D shows up as the last, rather than the first, panel. This is an innovation by the academy in its organiza- tion of the workshop. I suspect it is a harbinger of the fact that many of our panels, including my own, may urge Uncle Sam to stay out of R§D. At the summer planning workshop, I saw this writing on the wall when the discussion kept returning to Robert A. Charpie's report—the one that he did almost l0 years ago. Several who were present there kept emphasizing how well it hit the mark. That report, you know, em- phasized that the way the federal government could help most was to decrease or improve regulation, to improve taxation, to get better finan- cial and accounting structures and not to do federally selected and sponsored research and development. Then our keynote speaker started out on that theme. He suggested that we should use current ideas, which I suppose means that we do not need to develop new ones with federal R§D. Charpie followed right through, saying, "The government itself should not be in the innovation business." Our industrial speaker said that specific government action is not needed. Bill Saunders gave me some hope when he said government action was needed; but then, later on, he said: "It is not a research effort at all that I am talking about." I began to think we would have a fairly demoralized panel, and I thank Allen Puckett very much for saying $0.38 billion for DOT research is not very much at all. But there are, in fact, many very important questions that are being raised about how much is the right amount of federal involvement in R§D in transportation and where it should be spent. Court Perkins said in his welcome that he thought we were clever to narrow down the problem and to avoid generalizing too much about innovation. Instead he suggested that we really go ahead and look at it l55

in the transportation setting. As I have thought about this job, it seems to me that, in fact, we may not be narrowing down enough. I see at least five different sectors of government transportation in which we could have very different research and development policies and very different research and development activities. These differences relate to something that has been mentioned here several times, and that is the widely varying federal involvement in the different aspects of the transportation business. In this regard, it seems to me that there is a spectrum. At one end of the spectrum there are organizations like the DOD and the Coast Guard, and I think that is the full set, where all of the regulation, marketplace, and use decisions are made for the most part within those organizations. Coordinated efforts are not needed. So there is a real sense of accountability. If you do research and development and it goes no place, then you can start to figure out why. In these depart- ments you can find out much more easily what the operators want, although even then it is tough sometimes. At the other end of the spectrum, in areas like pipelines and barges, so far there has been very little, almost no, federal involve- ment. We have three sectors in between. One sector involves organiza- tions like NASA and the FAA, who in many ways are their own users; on the other hand, when they are part of a larger system, they have to interact with that system. Certainly, the FAA traffic control system has to work with pilots, airplanes, and the air transportation companies. Then in the middle of the spectrum there are causes such as mass transit and rail where the government has a great deal of economic and legal leverage. Allen Puckett mentioned the grants, and, of course, this implies that there must be some sense of accountability. Finally, there is the automobile, where the government makes virtually no marketplace decisions regarding who buys and sells what vehicles. On the other hand, they do have a tremendous impact and a growing impact through regulation. These are the five areas of interest to us. As our panel considers research and development programs and policies, we recognize that this is quite a broad spectrum, and a sound policy in one of these areas could be an anathema to another. I have analyzed the two papers that were commissioned for our panel, added some of my own thoughts, and put down what I think are ten issues that we should be considering. I will go over these very quick- ly and give you briefly some of the related recommendations contained in our two papers, as well as some of the issues that I thought were left out. The first issue involved the question of what can be done to improve our understanding of national transportation as a system, both currently and in the future. If we could get a good understanding of the national transportation system and the possibilities for the future, what impact would that have on our technological contributions to inno- vation, recognizing that technology is only a part of—maybe even a small part of—innovation? Along these lines, Edward Morlok feels very strongly that we need something that he calls a mobility assessment. In support of this, he l56

said, "We know so little about how the whole system works, and how the different parts come together." He proposes that an important function of DOT, which would not necessarily be very expensive in terms of dollars, but certainly would be in terms of intellectual horsepower, would be to have what he calls a basic research program in transportation. One could also describe his proposed activity as a very deep market analysis that would look at questions such as the role and the need for trans- portation in society, alternative technologies to meet those needs, alternative organizational arrangement at all levels for transportation, and the impact of transportation on society. The second issue is, "Generally, what is the appropriate federal role in research and development as contrasted to that of the local government, universities, private industry, and others?" As I suggested earlier, it may be that there are five different federal roles, depend- ing on which transportation area is served. Larry Goldmuntz implies this in his paper as he concludes that we ought to put our federal R§D assets where they are appropriate for federal involvement. He stresses air traffic control and surface transportation for a variety of reasons, and he says we ought to get the federal government out of automotive R§D. The Morlok paper, among its recommendations, talks about a cataly- tic role that the federal government could play. He proposes that we should establish institutes that would help integrate, that is, orche- strate, the efforts of universities, private industry, and other govern- ment programs. The third issue then involves the questions, "Are there, in fact, promising areas for federal funding of research and development in trans- portation?" It seems to me that most people agree that there should be some federal role in some of the basic technologies supporting transpor- tation. That has been mentioned several times today. Some of us feel that in air traffic control and surface transportation, an important role could be played. And, of course, there are some very contentious areas such as aircraft R§D. For a while there was a proposal that the government should get involved in developing new wide-body jets. Of course, there is the SST issue. I have already mentioned the role of the automobile. The fourth issue is, "Do we adequately follow the foreign competi- tion or foreign markets in transportation?" In Goldmuntz"s paper, there are some observations about German, French, and Japanese technolo- gies, and he concludes that they may be ahead of us in some important areas. If they are, that could hurt us in transportation, in our own transportation industry, and in our balance of payments. "Can we improve the means for funding of R§D?" is the next issue. Morlok proposes a set-aside tax that would be relatively quite small and would guarantee continued funding. There have been proposals made that we could jointly fund with industry. Another idea is that, with grants, we could combine R§D funding with some of the capital money and get better results. If there is to be federal involvement, another very important issue is, "What steps will improve the communication between developers, buy- ers, operators, and users in order that the developments may be more relevant and acceptable?" We have talked about the many people who are l57

involved. The best statement that I have seen of this subject was in Goldmuntz' paper. He says of one large transportation sector: "The user is not the buyer, the buyer doesn't pay for it, the payer usually doesn't buy or use it, and the operator is a professional who doesn't use, pay for it, or buy it." In that kind of a situation, it is very tough to try to discover what technology or innovation will prove successful, operational, and economical in the end. In fact, in Goldmuntz' paper, he cites many specific programs in which those kinds of disconnects and lacks of accountability have been at the heart of some major programmatic prob- lems. There have been successful solutions. He mentions a recent study of the FAA in which the FAA has retained an independent group to get together representatives of all the various players in this game and to look at some of the major mission thrusts that could be made. It seems to me that many of Morlok's ideas on the institute, on the mobility assessment, and on the Department of Transportation market research program could be a very important vehicle for getting more and better communication between these different groups. A seventh issue is, "How can we improve the collaboration between government and industry where each is sponsoring research and develop- ment in the same or closely related areas?" Our previous speakers have talked about some of the hostile behavior that some government people seem to show toward industry; and, on the other side of the coin, con- sider the low confidence that industry generally has in the technical ability of individuals and groups in the government to make technical and economic decisions. If, in fact, we are going to continue and strengthen our R§D pro- grams- -for example, those in basic research and technology underlying transportation—! know from my own experience in DOD that it is very difficult to make sure that those kinds of programs are relevant to the later system choices. This is going to be a very difficult thing to do. An eighth issue involves the question, "How can DOT's management be improved in the area of research and development?" When I heard Ward Haas say earlier that good innovation in technology means that the organization has to do such things as decentralize, make something as small as possible, short-circuit, and consistently fund, I began to get pessimistic about whether, if the federal government is involved, we can, in fact, get a good management of innovation. In fact, two of Goldmuntz' three major recommendations address the management question in DOT. In one, he emphasizes the need to stream- line the technological approval process and gives an example of some problems we have had there, and, in another recommendation, he would depoliticize DOT. He points out that we have had five secretaries of transportation and five UMTA administrators in the last l0 years and that in successive years we have found a real discontinuity, a reversal of decisions that makes any kind of progress in economic-technical development very difficult. We have to, under this organization question, consider the impact of the various modal administrations on technology development. You l58

may recall that when Bob Charpie looked at commercialization in the Department of Energy, his recommendation was that the department should be organized the way industry is, not by mode but by function; he felt that if this did not happen, if we did not have the organization struc- tured from the development, to planning, to marketing, and to production, that we would find that much less commercialization of the government- developed technology was taking place. Therefore we will have to look into that question of the modal administrations, the strengths they have, and some of the barriers they present to innovation. The ninth issue is, "Does the somewhat fragmented organization of the federal government act as a barrier to transportation technology?" I think we are commissioned here by DOT, but it is important to recog- nize that three other organizations are heavily in the transportation development operation business—NASA, DOT, DOE, in areas such as energy conservation, and EPA, in the environmental business. And, if we are going to talk about more basic programs, there is always the role of the National Science Foundation. Whereas we may need consistency in regulation, we fortunately do not need consistency, or not as much, in R§D. On the other hand, it is important to see that there is adequate communication between these various programs, that the right agencies are used in the right way; this raises the question, for example, about NASA. Is it possible that we want a national transportation R§D plan? I, myself, would be very pessimistic about the ability of the federal government to put one together. Then the last issue is, "Would changes in federal regulation, fund- ing arrangement, taxation, or others of the Charpie-like rules help to stimulate transportation technology in industry?" One of the finest parts of Morlok's paper addresses this question. He points out that because of this encrustation of rules very little change takes place, and that change is certainly not necessarily innovation, but that when there is very little change taking place, then it is an almost impossi- ble climate for any new technology to come in or for people involved to feel that they want it. And so he has made a recommendation, and I quote, "...it is imperative that a policy of the Department of Trans- portation be to take an active role in insuring that laws and regula- tions of the federal as well as other levels of government be written in a manner that is conducive to socially desirable forms of innovation." That is our agenda. I will repeat what the other panel chairmen have said, and that is, the final report may be organized in a comple- tely different way. Also, I like Bruce Old's comment that we need to find out who is on the other end of the telephone. One of the advantages that our panel will have is that we know pretty well who is on the other end of our telephone. l59

TECHNOLOGY RESEARCH AND DEVELOPMENT POLICIES OF THE FEDERAL GOVERNMENT AND TRANSPORTATION INNOVATION BY EDWARD K. MORLOK This paper is intended to provide background information and ideas for the panel session entitled, "Technology and R§D Policies to Stimulate Innovation." It deliberately takes a rather broad view of the subject This is primarily because a wide variety of policies affect innovation, research and development, and the deployment of transport technology. Furthermore, we contend that some fundamental changes in federal poli- cies are necessary to stimulate innovation in transportation that is socially desirable. Many of the ideas and proposals contained herein will prove quite controversial; it is hoped that they will stimulate discussion and provide a basis for specific recommendations to DOT re- garding policy changes and areas requiring more detailed investigation. SOME CONCEPTS Before we begin the discussion, it is necessary to define a few impor- tant terms and concepts with respect to both innovation and transporta- tion. First, by innovation we mean the creation of a new product or service, or creation of a new process by which to produce that product or service, or a combination of the two (4). Innovation is differenti- ated from invention by virtue of the application or use of that new product, service or process. Without the application, the new ideas do not constitute an innovation, but rather simply an invention. Or to state the distinction more simply, "the difference between the processes of invention and innovation is the difference between the verbs "to con- ceive and to use." (ll, p. 2). Thus, the term "innovation" is certainly not synonymous with tech- nological research and development. Innovation refers to a process. This process typically consists of four steps (this being taken from some unpublished writings of Herbert Holloman): l. Creation of the new idea (product or process). 2. The research and development necessary to make that idea reason- ably practical or applicable. 3. The -refinement of that idea into a practical one, based upon initial application and testing. 4. General adoption and dissemination of the idea. This description of the process in four steps is really one that reflects the view of the innovator, but there is a broader view that encompasses the assimilation of the idea into society and the further impacts of that assimilation. The process of innovation is really part of society's general process of adaption to change, whether that change be in resource l60

availability, in the needs or desires of people, or in other external conditions. As stated in the Department of Commerce report on innova- tion (ll, p. 2), ...invention and innovation encompass the totality of processes by which new ideas are conceived, nurtured, developed, and finally introduced into the economy as new products and pro- cesses; or into an organization to change its internal and external relationships; or into a society to provide for its social needs and to adapt itself to the world or the world to itself. These terms need some elaboration with respect to transportation. In the context of transportation, the product can be conceived of as a combination of the price and other characterisitcs of the transporta- tion service that are important from the standpoint of the customer—a traveller or shipper of freight. These other characteristics include such items as travel time, comfort, likelihood of damage to goods, etc., and the collection of these is usually termed "level of service." Thus from the standpoint of a user (or buyer) of transportation, the product is described by its price and level of service. Since we usually speak of transportation as a service rather than as a product, for the remain- der of this paper we shall use the phrase "transportation service." An innovation could thus be the creation of a transportation ser- vice that is new in the sense that it has a combination of price and level of service that is different from preexisting forms of transporta- tion. Thus high-speed rail service (once implemented) represents an innovation, since it is higher in speed and perhaps different in other characteristics from prior rail service. Similarly, the introduction of very rapid package or mail delivery services, such as Federal Express or the U.S. Postal Service Express Mail, also represents an innovation in the sense of a new transportation product or service. A useful way to think about a transportation service innovation is by means of a figure, such as Figure l, in which level of service is on one axis and price is on the other (7). Existing forms of transporta- tion are indicated there, along with a new or innovative form of trans- portation, which is indicated by a new point in this level of service- price domain. The second aspect of innovation relates to creation of a new pro- cess for producing an existing or new transportation service (in the level of service-price sense). An example of a process innovation would be the automation of rail rapid transit trains, which did not create a new service (since it did not alter in any significant way the quality of service or price of an existing transportation mode) but rather simply provided an alternative way to produce that service, using a higher degree of automation and less labor input. This change in the process of producing transportation represents an innovation. Of course, if there are substantial cost savings resulting from the application of such automation, then it might also change the transportation service in the sense of lowering its price. Many innovations in transportation consist of either one of the other of the two types of changes, as indicated by the example above, l6l

Price, $/shipnent A New express parcel service Preexisting "modes" JS. Level of Service e.g., Travel Time, hr. FIGURE l Description of innovations in transportation products or services in terms of price and level of service. Hypothetical example of a new transportation service or mode option for the shipment of parcels between a specific pair of cities. The new service has a price and service qualities that differ from previously available options. l62

while other innovations represent both new transportation services and new processes. For example, the introduction of air travel would repre- sent an innovation consisting of both a new process, that of powered flight, and a new service in the sense of a new level of service and price combination. An important question that remains is exactly how technology relates to these concepts of service and process and thereby to innovation. Technology refers to the process by which resources can be transformed into products (or services) that are of value to mankind. Our techno- logical capabilities in carrying out this transformation can be describ- ed by the range of products (or services) it is possible to produce and the resources used or costs (in its broadest sense, including environ- mental costs, for example) associated with producing them. One portra- yal of this might appear as in Figure 2. Here, for simplicity, only one measure of cost is used, and the product is homogeneous, so only the quantity produced is variable. At any point in time, the available technology might permit producing this product at the cost given by the line CD. As a result of development of new technology, it may be possi- ble to reduce cost per unit to the line FG. Turning from technology in general to transportation technology in particular, the capability to produce transportation can similarly be described in terms of costs and the nature and quantity of the product (or transportation service). The product characteristics are described by what was defined earlier as the level of service. The cost in general will depend on the amount of usage of the transportation system (because the amount of usage affects the amount of fuel consumed and because of congestion phenomena affecting the level of service, to men- tion two reasons), and hence there is an added dimension of usage. Thus the technology of transportation can be described by the relationship between cost, level of service, and usage, as shown in Figure 3. (This scheme for describing transport technology is developed and discussed more fully in reference 8, chapter l.) This conceptualization of transportation technology provides a framework for considering the effect of technological innovations. The set of points in this level of service-usage-cost space that corresponds to the known processes for providing transport describes the current state of transport technology. These might appear as the surface A in Figure 4. This would encompass all the known variations in hardware and operations of the existing means of transport and would include all the different "modes." A technological innovation could result from two types of change. First, technological developments may enable the provision of a level of services that was heretofore unavailable. This extends the range of possible transportation service qualities as indicated by area B. The corresponding cost is given by the extension of the cost surface labeled B'. The second type of change is a reduction in the cost, as indicated by the shift of part of surface A to A1. Such changes could result from technological developments embodying new hardware, or new ways of operating that hardware (operations plans), or both. The previously mentioned example of the innovation of express, l63

Total Cost 0 L Quantity FIGURE 2 Technology described by the relationship between cost and quantity of a particular product. l64

Cost A Level of Service FIGURE 3 Characterization of transport technology as the relationship between cost, level of service, and usage. l65

Usage Level of Service FIGURE 4 Technological innovations characterized by changes in the cost-usage-level of service relationship. l66

guaranteed delivery parcel service resulted primarily from operating available hardware in a new way. On the other hand, a new high-speed rail passenger service typically would involve primarily hardware changes from more conventional rail passenger service. In this discussion of technology, price has not been mentioned. The reason is that the price charged a user does not necessarily bear any particular relationship to cost or other characteristics of the technology. It is determined primarily by institutional considerations. Of course, in our society most institutions are required to generate total revenue from users at least equal to total cost, although many transportation organizations are heavily subsidized so that prices can be set far below monetary cost. What then are the roles of technology research and development in transportation? Clearly, these are (l) to expand the range of level of service that can be provided,and (2) to reduce the resource costs. These bear a brief discussion. Expansion of the level of service options can occur in many ways. One is to develop what might be considered an entirely new type of trans- port technology, such as the aircraft. Some such new technologies do expand the range in ways that appear to have been useful and beneficial, as in the case of aircraft. Others, though, while new and different, meet with little acceptance and hence are not so beneficial, an example being the monorail (at least up to the present). On the other hand, many technological innovations in transportation that embody a new level of service require only modest or incremental extensions of exist- ing technology. The express package services are one example, and the dial-a-bus concept in urban transit is another. It is important to note that all service innovations in transporta- tion do not require any underlying technological innovation. A service innovation may be simply a change in price. Special reduced fare plans represent one example, and the lowering of air fares in some markets as a result of increased competition and altered regulations represents another. The reduction of costs also has many aspects. It is usually dis- cussed in terms of reducing the monetary cost of production, but this is only one aspect. Cost is used here to mean all resources used in the broadest sense, and therefore cost refers also to resources used for which there may be no monetary payment, such as air pollution and noise (where the resource "used" is clean air or quiet). SOME OBSERVATIONS Although the preceding presentation of definitions was somewhat tedious, it will prove useful immediately because it provides the basis for a number of points related to innovation and R§D in transportation. An extremely important point—and one that is often missed—is that there has been and probably continues to be a very substantial amount of innovation in transportation, despite all the rhetoric to the con- trary. One need not look far to see this. Since World War II we have l67

witnessed innovations such as the introduction of jet aircraft, the rapid expansion of long-distance pipelines, the emergence of long- distance trucking in its many forms (private, contract, common, etc.). and the introduction of rapid, guaranteed-delivery parcel services-- all of which represent both new transportation services and new pro- cesses. Also very recently, we have seen considerable innovation in urban transport, some involving new services and processes, such as dial-a-bus and Personal Rapid Transit (PRT), others providing innova- tive services but not involving a very substantial process innovation, such as transit services targeted to the needs of particular groups (e.g., for elderly and handicapped, high-quality commuter service, commuter clubs), and still others consisting of process changes that affect only resources used (e.g., automated rail rapid transit, rubber- tired rail transit). Furthermore, the R§D underlying new inventions, and hence potential innovations, continues. Much of this is process oriented and often designed to reduce costs, such as work on improving fuel efficiency found in every mode. Other efforts portend changes in service, too, such as the railroad industry's efforts in car routing and control and the computerized methods for real-time monitoring and control of urban street networks (already implemented with respect to a few control variables on a small scale in a handful of cities). Not all efforts are directed at refining what might be thought of as existing forms of transportation; some research on what might be termed entirely new systems is also underway. In particular, there is some work on the movement of merchandise freight in capsules in liquid or gas pipelines. Interestingly, to the best of my knowledge, the only R§D work on this in the United States is being undertaken by private firms, except for one small, preliminary study of economic feasibility of the so-called "solids pipeline" sponsored by DOT (l2). Private R§D work has reached the stage of a short demonstration line to test the equipment. Prospects for this technology are difficult to judge, but even a former administrator of the Federal Railroad Administration has referred to solids pipelines as the '"sleeper1 in freight transport technology" (6, pp. 54-55). Perhaps DOT has done so little in connec- tion with solids pipelines because there is no home for such work; there is no modal pipeline administration, and of course such technology would be viewed as a threat in the context of other modal administrations. It has also been argued that organizations currently engaged in transportation will concentrate their R§D efforts only on innovations that represent evolutionary changes in the technologies with which they are concerned. An implication of this, some argue, is that an entirely new form of transport would never emerge from such an evolutionary pro- cess. This is then taken as evidence of a need for government-sponsored R§D in new forms of transport, if such new forms are ever to develop. It is not at all clear that new forms of transportation, representing very significantly improved transportation services (in the level of service-price sense presented previously) or reductions in resource needs, would not occur. One reason is that what might be viewed initi- ally as an evolutionary step in an existing form of transport may become l68

the basis for a major improvement in service and costs. An example is the replacement of propellers by the jet aircraft engine. Another reason is that one form of transport may emerge gradually from a pre- vious one that continues on its own development path. An example is the railroad, which today is radically different from road transport but which evolved from horse-drawn wagons on gravel roads in the eighteenth century, the first railroads being simply road wagons operating on plank paths or "tracks." Thus, given the freedom to alter technology, a new form of transport can emerge in an evolutionary way to become what is then thought of as a form distinct from its origins. Given the level of research, development, and implementation of innovations in transportation, care must be exercised so that changes in federal R§D policies do not damage desirable features of current efforts. In particular, care must be taken lest any increased federal efforts simply drive out private efforts and efforts at other levels of government. ROLE OF GOVERNMENT It has sometimes been argued that national governments really have relatively little influence on the pattern of innovation in society. For example, a l97l Organization for Economic Cooperation and Develop- ment report concludes (l0, p. l38): The available information also shows that the main agents for the creation and application of scientific and technological innovations are the universities and industry. Uncertainty, change, the need for competition, flexible structures, rapid decision making, and being close to technological and market developments, all imply that technological innovation is more likely to flourish in a decentralized and pluristic environ- ment. Thus, government roles in the innovative process, although important, are not determinant. This statement is perhaps a bit misleading, for it really refers primari- ly to the direct influence of government actions on the amount and character of innovations. Clearly, the influence of government extends far beyond its direct actions, because governments create the framework within which other organizations act, and therefore will influence the direction and amount of innovative activity of other organizations. The federal government could influence innovative activity of others through a variety of channels, including but not limited to direct in- volvement in steps of the innovative process, incentives to innovation through taxation policies, regulations, and policies regarding the creation and dissemination of new knowledge. The role of the federal government in transportation innovation is somewhat unique in comparison with other sectors of the economy, because of the direct involvement of government in the provision of transporta- tion facilities and services. Most transportation facilities are pro- vided by various levels of government, usually with considerable federal involvement in financing such facilities, sometimes from user charges l69

and sometimes from general taxes. Along with such funding there is con- siderable influence over planning and design, pricing, level of service, and other aspects of such facilities. Also, most of the transportation system that is not directly financed by the federal government--mainly privately owned carriers—is subject to regulation that is in addition to the general regulation of business enterprises. Such regulation is mainly by federal and state agencies and includes control over entry and exit from service, prices to be charged, and in some cases the level of the service. This involvement of federal government creates many ways in which it can influence the speed and direction of techno- logical innovation in addition to policies related to R§D in general. The ensuing discussion of possible federal policies toward research and development in transportation emphasizes the relatively unique in- volvement of the government in transportation. Thus the discussion touches only slightly on general policies toward R§D and innovation in general, such as in the areas of taxation or patent law. These areas are covered very well in existing literature(2,5,9) and an attempt to include the essence of such studies would be an injustice to such reports and make this paper overly long. Hence we shall focus rather specifically on "transportation" technology and R§D policies. This will be in the form of a discussion of major areas of policy, each of which will contain one or more specific policy recommendations. Where Innovation is Needed One of the most critical requirements is for information on where inno- vations are needed in transportation, or where innovation will produce benefits—to users of the transportation system or to others impacted by it. Such knowledge is of central importance as a guide to govern- mental policies related to technology R§D directly and to other policies that influence the adaption of the transportation system in general. Innovations can be beneficial because they provide a new type of transportation service in the level of service-price sense discussed earlier or because they employ new technology that reduces the societal costs or negative impacts associated with that transportation. Identi- fying where innovations will be beneficial requires attention to both of the possibilities. Turning first to "new" kinds of transportation services that might be offered, there would seem to be two primary means of identifying what types of innovations are likely to be beneficial. The first of these involves surveying users and potential users of the transportation system to ascertain the degree to which they are satisfied with the range of services (or "modes") now available and their current needs or expectations regarding the desirability of alternative types of service (again, in the sense of level of service and price combinations). Such a survey would in effect provide an assessment of the adequacy of the transportation system from the standpoint of performing its primary func- tion, that of providing mobility. Hence this survey can be termed a mobility assessment. l70

It is perhaps amazing that while transportation (and other activi- ties) is monitored from the standpoint of impacts other than that of its function, particularly environmental impacts, it is not now moni- tored with respect to how well it performs its primary function. Yet how well the transportation system meets the mobility needs of the nation is surely equal in importance to its potential negative environ- mental impacts. What is needed is a continuing assessment of mobility. Such a survey will be difficult to construct, but should provide sub- stantial benefits in terms of indicating where problems exist in the transportation system. It is important to recognize that this assess- ment should address the adequacy of the system from the standpoint of the user and should not be an assessment against arbitrary engineering or other criteria, a type of assessment that is often made but that is not truly indicative of the degree to which the system is performing its function adequately. In addition to providing guidance for techno- logy R§D, this assessment would also provide guidance for other policies and programs. The emphasis in the preceding paragraph on evaluation of the ade- quacy of the transportation system from the standpoint of mobility is not to imply that other viewpoints such as that of the environment are unimportant. Along with the mobility assessment should continue assessments of environmental impacts, and in some cases these should be expanded. This presumably would be accomplished in cooperating with agencies already concerned with environmental matters. As was pointed out earlier, to some extent this type of assessment is already common- place, as in the case of air pollution in metropolitan areas. Closely related to this survey of transportation system users would be an attempt to identify mobility and other problems that are likely to occur in the future. The basic purpose of this would be to antici- pate problems far enough in advance that corrective action, whether it would involve technological innovation or other changes to the trans- portation system, could be taken before the problem becomes damaging to society. It would attempt to assess future requirements for movement and then to determine the extent to which the system will accommodate those needs through its normal processes of adaptation and change. If that adaptation is insufficient, then this indicates a problem that requires corrective action. The corrective action could involve an attempt to develop new technologies or could be of another type, such as changes in regulations or funding. Some efforts are already being made along these lines, such as the National Transportation Needs Study, although existing efforts do not focus on anticipating future problems and attempting to design solutions. Another important source of information on types of innovations that would be beneficial consists of the many governmental organiza- tions that are already involved in planning and providing transporta- tion facilities and services. Primarily as a result of federal require- ments, there are agencies at various levels of government concerned with the planning of particular portions of the transportation system. These include metropolitan planning organizations at the local level, which are concerned with both passenger and freight transportation via all l7l

modes within metropolitan areas; statewide transportation planning organizations, which have traditionally focused primarily on highways, air, and to some extent, water transport facilities, although now they are extending their purview to rail transport and to other carrier services; and various federal planning activities, such as those related to the overall air transportation system and water transport. As part of the planning function of these organizations, future needs for trans- port are assessed, and attempts are made to develop plans for system expansion to meet those needs. At the present time, except for air transport, almost no considera- tion is being given to innovation within such planning agencies. Yet it is within these organizations that the need for and benefits of innova- tion could readily be assessed. Each planning unit presumably has developed some understanding of the need for transportation in its area of concern, and would be particularly aware of present and emerging problems, such as cost escalation or a lack of mobility among particular groups. A natural requirement or role for these organiza- tions would be to provide information on the types of innovations, both service and process, that would be beneficial to the areas with which they are concerned. Since so many of them operate under federal guide- lines at the present time, the mechanism already exists whereby these agencies could be required to produce statements of innovation needs and priorities on a regular basis. This leads to the first set of recommendations for DOT policy re- lated to R§D programs: The federal government, in cooperation with state and local governments as appropriate, should assess the per- formance of the transportation system at regular intervals. They should consider: o adequacy in meeting the needs of users. o viability of the organizations providing transportation facilities and services. o impacts on development, resource use, pollution, etc. A similar effort should be undertaken with respect to anticipating future problems—both near term and long term-- considering the likely natural adaptation of the transporta- tion system to changing conditions. Such a survey and studies would provide the basis for identifying and prioritizing problems in terms of severity and time frame. This emphasis on understanding what types of innovations are needed, or might produce substantial benefits, is intended to create what might be termed a benefit orientation to innovation, whether that innovation will be developed essentially through government programs, or by the private sector, or by a combination of the two. Unfortunately, many government programs concerned with technology research and development, especially in surface transportation, seem to have been motivated primarily by the realization that a particular form of transportation could be developed to the point where it is technologically feasible, but with little or no regard for whether that technology would in fact create net benefits greater than those associated with the existing l72

technology that it was designed to replace. Examples of this include many of the efforts to develop new types of urban transit technology and much of the R§D effort in high-speed, intermodal intercity freight systems. This is of course not to imply that innovation in these two areas might not be beneficial, but simply to point out that much of the current research and development does not seem to have been directed in ways that would produce substantial gains. To correct this, it is recommended that: Priorities for federal research and development efforts be based on consideration of (l) the role that innovations resulting from such R§D would play in alleviating current and anticipated future problems, and (2) the enhancement of the transportation system and its effect on the quality of life resulting from such R§D-associated innovations. The R§D Program Another important issue is what type of R§D program the government should have. A general conclusion that runs through much of the inno- vation literature is that the proper role of government in R£jD is to ex- tend the frontier of knowledge that is generic to a particular area, but to leave to the private sector the selection of specific technological innovations to be developed and brought into the market. This conclu- sion rests on a number of characteristics of the innovation process, in particular, the uncertainty of success, the need for firsthand knowledge of the potential markets as well as the feasibility of producing a particular new product, and the need for entrepreneurial skill, which are unlikely in governmental bureaucracies. Also, an important consi- deration is the fact that not much general or basic research would be undertaken by anyone, or not on a very large scale at least, without government support. This conclusion that the most effective role of government is in basic research seems to be so widely accepted that it would seem to be one of the most important policy guidelines for govern- ment R§D programs in transportation. However, again on this matter it is important to recognize that there are unique features of transportation that can modify these con- clusions about the roles of government and other organizations. Agencies of various levels of government are the buyers of many transportation system components, especially in the highway, transit, and air modes. As a result there are many impediments to R§D by suppliers of these com- ponents. R§D generally is not an allowable cost for which the govern- ment will pay. Also, some supplier industries may be so fragmented that no one firm could undertake any significant amount of R§D, as in some areas of construction. As a result, it may be necessary and appro- priate for the government itself to undertake R§D that pushes much closer to readiness for application, or to directly fund applied R§D in the private sector. Recognizing this caveat, though, the general policy guidelines would seem to be: l73

sponsored by the federal government should be directed toward providing the knowledge and results that are necessary for private firms or publicly owned transportation agencies (e.g., state road departments) to refine and develop the new concepts to the point where implementation is possible. In general, every effort should be made to provide the bases for alternative technologies, rather than focusing on one approach, allowing the market to decide which approaches will ultimately be used. Also widely accepted is the critical role played by basic research in technological innovation. For example, in a study of "five econo- mically and socially important civilian innovations," it was found that of 34l key events that led to these innovations, approximately 70 per- cent were the result of "nonmission research," essentially equivalent to what we normally think of as fundamental or basic research, 20 per- cent were "mission-oriented research," and l0 percent were the result of specific development and application work (l, p. 84). While the fraction of key events behind any innovation that results from basic research will of course vary considerably among innovations, the impor- tance of fundamental knowledge seems quite clear. One of the basic problems in innovation in the transportation field is that the state of fundamental, generalizable knowledge is very limit- ed and fragmentary at best. Most of the knowledge that has been genera- ted is the result of rather specific application-oriented efforts. These are dominated by attempts to refine existing technologies (e.g., better materials, improved design guidelines) and by attempts to better project future travel as a guide to investment using conventional tech- nologies. There is a continuing emphasis on research on specific mis- sions and immediate application, with only a tiny fraction of the Depart- ment of Transportation budget being devoted to basic research in trans- portation. If innovation in transportation is to be fostered, a program of basic research in transportation must be undertaken. Responsibility for this rests clearly with the Department of Transportation, for it is the only organization with the broad, long-term view and the financial resources to undertake such a program. Thus a recommendation is as follows: The U.S. Department of Transportation should undertake a program of basic research in transportation, which would com- plement and support its mission responsibilities and provide the knowledge base on which future innovations in transporta- tion can be developed. This program should include research on (l) the role of and need for transportation in society, (2) alternative technologies for meeting those needs, (3) alterna- tive organizational arrangements for supplying transportation, and (4) the impacts of transportation on society. Probably the most appropriate mechanism for undertaking such a research program would be a program of support to key universities and organizations concerned with basic research in this field. To be effec- tive, such support must give wide latitude to the nature of problems l74

studied, must be sufficient to support units of substantial size and diversity of talents at each institution, and must provide continuing funding over an extended period. Such a program would be comparable to the support for basic research in many areas of physical science and mathematics by the military in the two decades after World War II, a period in which there was tremendous advance and innovation in those areas. It has been observed that with the reduction in emphasis on military matters and the increase in attention paid to civilian prob- lems, one of our society's major problems is to reorient basic research toward the more important civilian problems and away from areas that have application primarily in the military. (This is discussed at length in reference 5.) Transportation is clearly one such area. Such a program of basic research in transportation would be in addition to the current research programs of the Department of Trans- portation, which are primarily within the several modal administra- tions. Current programs include technology research and development within the sphere of each mode, and for the most part this research seems quite adequate. The major exception is, of course, pipelines, since that mode is not represented in the department by an administra- tion, nor is it covered by a separate agency (as in maritime transpor- tation) . Technical Institutes Related to these considerations is the coupling of the results of R§D with potential users. A common suggestion is for this to be the func- tion of research institutes, which would be capable of conducting some of their own research and development (falling between the basic research typically conducted at universities and the applied R§D work typically undertaken in private industries). These institutes would be charged with the responsibility for working closely not only with governmental organizations involved in transportation but also with private sector organizations such as carriers, vehicle manufacturers, and construc- tion firms. Another function of these organizations could be to assist entrepreneurs who are developing new products or services. Such assis- tance could range from technical advice to helping to identify potenti- al users, financiers, etc. Organizations that perform some of the functions of such institutes already exist. These are in the research units that were formed at the state universities to assist state road departments in the construc- tion, design, maintenance, and operation of highways. Some of these have expanded to encompass other areas of transportation, but rarely with the close coupling to the various external organizations evident in the highway transportation areas. The tremendous strides made in highway design and operations attest to the value of such coupling, although of course it required substantial, continuing funding. But there are significant differences between the organizations involved in the provision of highways and those in other areas of transportation, so the same form may not be applicable elsewhere. However, the poten- tial benefits are so high that: l75

The DOT should explore the appropriateness of creating technical institutes where the primary function would be cooperative research and development involving various levels of government and private industry. R6D Funding Programs such as those sketched above will not be inexpensive, and it will not be easy to obtain necessary funding in this period of fiscal restraint. A separately budgeted research program would elicit con- siderable opposition, and scrutiny of each element would undoubtedly lead to elimination of many promising lines of inquiry. For these reasons, it would seem prudent to associate the research with major DOT program areas rather than have it appear as a separate, very vul- nerable, item. An approach that has been used quite successfully in the past in highway programs is to allocate a small fraction of program expenditures to research. Even though the fraction was quite small, l l/2 percent, the results in that field were very significant and beneficial, as was pointed out previously. A major difference between that highway research program and a general transportation research pro- gram would be that the general program should encompass all forms of transportation and should reflect the needs of users, other impacted groups, and transport technology and organizations in general instead of focusing on research that would necessarily fit into the purview of one or more modes. Therefore, on funding it is recommended that: To insure adequate funding of R§D, it should be DOT policy to set aside a small fraction of all expenditures for research and development. This set-aside should be distribu- ted between (l) R§D directed toward evolutionary improvement in that area of transportation (mode, organization, etc.), and (2) R§D directed toward providing the basis for major innova- tions and improvements in the transportation system in general. Barriers to Innovation A final important area for fostering technological innovation in trans- portation is the elimination of many barriers to innovation that exist as a direct result of the present form of federal (and other) govern- ment involvement in transportation. As was mentioned previously, transportation is somewhat unusual in that the government is heavily involved in regulating many transportation activities and also because the government is involved in funding many such activities. With respect to regulation, it seems clear that virtually no cog- nizance is taken of the effect of current economic regulation of trans- portation on the pattern or degree of innovation in the regulated in- dustries (.3). It is also clear that regulation is in no sense neutral with respect to technological innovation. In some cases, such as the airline industry, the rate of technological innovation may have l76

increased as a result of regulation in comparison to what it would have been without that regulation. In contrast, in the railroad industry it is equally clear that the current pattern of regulation has impeded many innovations that the carriers would have liked to have undertaken and undoubtedly has created a climate so difficult for much innovation that many are not even considered. Some change in this climate will undoubtedly occur as a result of current efforts at regulatory reform, but the reforms seem to be based primarily on considerations other than innovation, and therefore the result may still be a regulatory process not in consonance with innovation. Regulatory agencies must begin to consider the effect of their rulings on the propensity and ability of the regulated carriers to inno- vate and adapt to changing conditions. Changing habits or regulatory agencies will undoubtedly take many years to bring about, but certainly the change can be accelerated by specific requirements for such agencies to consider the effect of their decisions on innovation, in effect requiring them to produce innovation impact statements as part of their decision process. Another type of regulatory change that would drastically improve the climate for innovation would be to give providers of transportation services much greater latitude for experimentation with new types of services and processes for delivering them. Clearly, there would have to be some limit on the amount of service that could be involved in an experiment or the entire regulatory process would cease to exercise con- trol over the system and would become meaningless. But at the present time the regulation seems to be so rigid as to preclude trying many promising ideas, and unless they can be tried, they will never become innovations. This freedom to experiment on a limited scale would ideal- ly apply not only to carriers already performing a service but also to new entrants into the field, so that new forms of service, new institu- tional arrangements, etc., could all be tried. The second important area of government involvement is that of funding many transportation facilities. Highway facilities are largely funded through a hierarchy of charges or taxes collected from highway users by the federal, state, and local governments, most of which is spent by state and local agencies to construct and maintain the highway system. Other services such as urban public transportation and Amtrak are financed by federal and local monies collected from general taxes and revenues from users, while others such as the waterway system are funded primarily from general taxes only. These mechanisms often are quite rational from the standpoint of raising money and then providing that money to those who need it, but such mechanisms often are deficient from the standpoint of encouraging efficiency and innovation in socially desirable ways. Perhaps one of the most notorious examples of this is in connection with the funding of the Interstate System, in which each state is permitted to build up to a certain number of miles of inter- state highway. As the Interstate System is nearing completion, states are realizing that it is in their selfish interest to make the remain- ing few miles as expensive as possible, in order to maximize the influx of "federal" monies into their states. This creates a disincentive to innovation that would reduce the cost of such programs. l77

Less dramatic, but perhaps of more significance given the amounts of money involved, are the consequences of federal funding mechanisms for other parts of the transportation system. For example, in the case of more typical urban transit or highway projects in which the federal government pays 80 percent of the capital costs (the state and metropolitan area paying the remainder), there is the temptation to maximize expenditures so as to maximize the amount of federal money flowing into the area. After all, $4.00 or more of "federal" money is attracted to the region for every $l.00 of local money. Furthermore, there is a natural hesitancy of those in the political process to ex- periment with any innovation, for these have a higher risk of failure than traditional approaches. Few persons are criticized for taking the safe course of traditional technology even if an innovative approach might prove more beneficial, but criticism will be loud and clear if one initiates a failure. Also creating disincentives for innovation is the form of federal funding of operating losses on urban transit. Current laws provide for the matching of local funds, with little or no regard for the effi- ciency with which funds are expended. This, in conjunction with the politicization of key managerial functions (fares, areas to be served, routes, levels of service, and management posture in labor negotiations), and the monopoly position of most public transit organizations, creates a climate that discourages much innovation. Added to this are federal requirements for capital project funding that protect labor to the point of perpetuating outdated staffing practices. This is a dismal setting for innovation indeed. Thus substantial disincentives to innovation are built into exist- ing federal programs in transportation. This has a direct bearing on technology R§D, for if there is little likelihood for implementation of new or improved technology, there is little point to developing it. It is analogous to pushing on a rope; unless there is a pull at the other end, nothing will happen. Overcoming these disincentives will not be an easy task. One solution is to increase the possible gains to the local area resulting from innovation. The disincentives inherent in current funding might be eliminated by substitution of some type of grant funding for match- ing funding. In the grant funding there would be a ceiling on the amount of money that might flow into an area for transportation projects. Such funding would create an incentive to maximize the benefits to the region from the expenditures, and innovations that promise particularly large potential increases in those benefits presumably would be more likely to be tried. If an innovation were successful in one location, then the uncertainty or risk in other situations is reduced and it is more likely to be applied elsewhere. This type of funding change in conjunction with increasing sensi- tization of state and local transportation agencies to the prospective benefits of innovation (through such means as requiring innovation priority statements, as mentioned above) should go a long way toward improving the climate for innovation. If federal efforts at innovation are shaped by local priorities for innovation, then the innovations l78

that are developed should be more in keeping with local needs, and the likelihood of implementation should increase correspondingly. Much of the foregoing has related to innovation in general rather than technology R§D in particular, but there is a direct connection. If there are substantial disincentives or barriers to innovation, then there is little likelihood of even very promising ideas being implemen- ted. In this case, research and development of improved technology is a futile exercise, for it is unlikely to be used. Moreover, efforts to develop an R§D program are likely to be opposed, for many will see it as a waste of government energies and resources, and justifiably so. Thus the climate for innovation has a direct bearing on the development of an R§D program. It is recommended that: Just as transportation R§D must ultimately be responsive to the needs and priorities of society, so must the private and public organizations that control the transportation system be receptive to innovations. Therefore it is impera- tive that a policy of the Department of Transportation be to take an active role in insuring that laws and regulations of the federal as well as other levels of government be written in a manner that is conducive to socially desirable forms of innovation. EPILOGUE This paper has approached the question of federal policies on technolo- gy R§D and to stimulate innovation in transportation from a rather broad perspective. It has focused on questions of the proper purpose and scope of the DOT R§D effort, its relationship to other organiza- tions involved in the provision of transportation equipment, facilities, and services, and its relationship to other governmental policies in transportation, such as economic regulations. It has suggested some policy guidelines in these areas, but in the spirit of the charge "to develop suggestions for more detailed analyses and evaluation" it has not attempted to carry suggestions to the level of specific actions or programs. The panel sessions will provide an opportunity to attempt to develop more specific recommendations to DOT regarding policy changes and questions that require further attention. l79

REFERENCES l. Abernathy, W.J., and V.S. Chakravarthy. Technological Change in U.S. Automobile Industry: Assessing Past Federal Initiatives. DOT-TSC-NHTSA-79-l3, Transportation Systems Center, report to the National Highway Traffic Safety Administration, by Lexington Tech- nology Associates, Lexington, Mass. Washington, D.C.: U.S. Depart- ment of Transportation, l979. 2. Advisory Committee on Industrial Innovation, of the Domestic Policy Review on Industrial Innovation. Review and Recommendations of the Policy Alternatives of the Public Interest Advisory Subcommittee, draft report.Washington, D.C.:U.S. Department of Commerce, l979. 3. Boyd, H. Inducing the Development and Adoption of Socially Effi- cient Automotive Technology. DOT-TSC-RSPD-78-4, Research and Special Programs Directorate. Washington, D.C.: U.S. Department of Transportation, February l978. 4. Gellman Research Associates, Inc. Economic Regulation and Techno- logical Innovation: A Cross-National Literature Survey and Analy- sis . 3 vols. A report to the National Science Foundation, Jenkintown, Penn., l974. 5. Gilpin, R. Technology, Economic Growth, and International Competi- tiveness. A report prepared for the use of the Subcommittee on Economic Growth of the Joint Economic Committee, Congress of the United States. Washington, D.C.: U.S. Government Printing Office, July l975. 6. Lang, A.S. "Demand and Supply: The Technology of Transportation." In The Future of American Transportation, edited by Ernest W. Williams, pp. 4l-57. Englewood Cliffs, N.J.: Prentice-Hall, l97l. 7. Morlok, E.K. Introduction to Transportation Engineering and Plan- ning. New York: McGraw-Hill, l978. 8. Morlok, E.K. An Analysis of Transportation Technology and Network Structure, Evanston, 1ll.: Northwestern University Press, l969. 9. National Academy of Sciences. Technological Innovation and the U.S. Economy. An Academy Forum held in Washington, D.C., November l4, l973. l0. Organization for Economic Cooperation and Development. The Condi- tions for Success in Technological Innovation. Paris, l97l. l80

ll. U.S. Department of Commerce. Technological Innovation: Its Environment and Management. Washington, D.C.: U.S. Government Printing Office, l967. l2. Zandi, I. et al. Transport of Solid Commodities via Freight Pipe- line. 4 vols. A report prepared for the Office of University Research. Washington, D.C.: U.S. Department of Transportation, July l976. TECHNOLOGIES AND R§D POLICIES TO STIMULATE INNOVATION BY LAWRENCE A. GOLDMUNTZ INTRODUCTION The practice at the federal level of stimulating innovation in trans- portation is faulty, lags behind that of our international competitors, and has diverged from theory. This results in serious consequences for the consumer, industry, and the competitive position of U.S. commerce. The stimulus for innovation is frequently discussed in terms of demand-pull or technology-push. In demand-pull, the usual strategy is to depend on involvement of the users of the innovation or their surro- gates, market-research personnel, to determine market needs. Under this strategy, the role of government is to remove obstacles to the free expression of market forces so that development of technologies will be appropriate and will satisfy operators and users. In the technology- push approach to innovation, one relies on the market intuition of technologists as to what users or operators, that is, the market need or want. The demand-pull strategy has been exemplified, at least historically, by the automotive industry. Technology-push is exempli- fied by NASA's programs. While the distinction between demand-pull or technology-push may be useful to analyze technological innovation, both are frequently necessary to introduce innovation into the marketplace. Successful innovation depends on knowing when each approach should be emphasized and when one might make a transition from one approach to another and which institutions to go to to make sure that each approach has a reasonable chance to make its contribution to the introduction of innovation into transportation. While there has been innovation in transportation, it has not been in general due to a thoughtful federal initiative. At times the federal role seems perverse: demand-pull has been used where technology-push would probably have been more effective and vice versa; the ability to respond to markets has been inhibited when the government itself if the customer; and, in some cases, modes have been pressured to accept inappropriate technologies and these actions have usually been publi- cized with elaborate performances by public affairs officials. l8l

If these comments seem harsh, consider the following: l. The Metro cars for the Washington-New York corridor were speci- fied and designed in l967 to operate at l75 miles per hour although the roadbed, catenary, and trains on parallel track would be destroyed at these speeds. The justification was to outperform projected Japanese trains, which were in fact not built. A high-speed air cushion vehicle from McLean to Dulles airport was proposed by a secretary of transportation to be constructed within l8 months, in time for Transpo l972 as a demonstration project. Technolo- gists questioned the schedule, economists questioned the market, OMB observed that it went from no place to nowhere. It was cancelled. 3. The Morgantown Project was to be completed in time for the President's daughter to make a campaign-oriented inaugural ride. Trouble-free service occurred much later. 4. Passive automotive restraints were mandated before there were field data on the performance of passive belts. A program to obtain adequate field data on air bag restraints proposed by one secretary of transportation, who then decided to reinvent the automobile, and then resigned, but for other reasons. 5. The Downtown People Movers demonstration was inaugurated by DOT, but an adequate technological base to provide options for city planners was lacking. The technological options made available to German city planners by their Ministry of Research and Technology are considerably more extensive and have been developed in a shorter time and with less funding than comparable U.S. programs. 6. Transbus specifications were changed by four successive secre- taries of transportation. The last change--made despite warnings of problems perceived by both operators and manufacturers and contained in an Office of Technology Assessment report—resulted in no manufacturer responding to the bid request. 7. A microwave landing system was first demonstrated in l946. A more refined microwave landing system was approved by the International Civil Aviation Organization (ICAO) in l977. Implementation is not yet in sight. Yet 50 percent of fatalities are associated with the approach and landing phase of flight. A constant concern of users of the air traffic control system relates to the length of time taken to complete and implement engineering and development programs. While all these programs are complex, and while capsule criticisms are sometimes glib, the thrust of this small sampling is correct: DOT has not stimulated innovation successfully. DOT STIMULATION OF INNOVATION IN AUTOMOTIVE TRANSPORTATION--WHERE IT IS NEITHER CUSTOMER NOR USER The proper federal role in transportation innovation should vary signi- ficantly from mode to mode. In automotive transportation the govern- ment's role over the last l0-l5 years has been to promulgate certain health and safety standards and, more recently, certain performance standards with respect to fuel economy. Prior to these l0-l5 years, the l82

government had essentially no role with respect to the automobile it- self, although it did have a significant role with respect to arranging the financing for highways and highway maintenance. Consideration is now being given to an expanded federal role in research as it applies to the automotive industry. There have been some minor and mostly unsuccessful federal research and development efforts in support of the automotive environmental and safety regulatory programs. For example, DOT's Research Safety Vehicle program and EPA's, then ERDA's, and now DOE's support of automotive power systems that might satisfy the requirements of the Clean Air Act and be fuel efficient have not successfully achieved their objectives. The Automotive Power Systems Program had expanded more than $50 million by l977 and has since been budgeted for much higher expendi- tures. The initial purpose of the program was to demonstrate automo- tive engine technologies that would meet the l975 emission standards established in the Clean Air Act of l970. The program in fact had little impact on any of the technological approaches that the automo- tive companies selected to meet the emission standards. The program concentrated on external combustion technologies rather than exhaust treatment. All the external combustion technologies turned out—in accordance with the automotive industries' predictions—to be signi- ficantly less efficient than internal combustion technologies. The Office of Management and Budget directed that program efforts should not be duplicative of automotive industry efforts. This seems quite reasonable. But this forced the program to look at precisely those technologies that the automotive industry found unattractive. This is not a recipe for efficient use of federal funds. There have been claims that this program could have been more successful if its designers had recognized that automotive technology was mature and federal support has to be at a more basic level in the R§D hierarchy in order to make a contribution and that basic support of catalyst phenomenon or combustion theory research would lead to a successful federal intervention. However, catalyst experts today make a good deal of money protecting their proprietary position, the experts are likely to be wary of the constraints associated with federal money, and the government should not support the inexpert. Furthermore, those technologies that might lead to low pollution levels and high efficien- cies—the Stirling and Stratified Charge engine—are probably more limit- ed by cost and manufacturing technology than by basic combustion theory. There is thus some question as to the effectiveness of federal support of automobile engine manufacturing technology—a subject more familiar to industry than government. Some say that EPA's Automotive Power Systems Program was initiated to forestall an even more expansive program that Congress was intent on legislating. Perhaps the Automotive Power Systems Program exemplifies a damage-limiting justification for R(jD investment, which is a novel wrinkle in economic theory. However, as serendipity would have it, evidently some burner technology—developed under this program for tur- bines—may be useful in boilers, and ceramic turbine blade developments, also supported by this program, may yet be developed and find applica- tions. l83

Some claim that federal automotive RfiD support is needed for those companies in the industry that are marginally viable. Automotive companies are marginally viable because of management problems and federal regulatory excesses, and it is hard to conceive how federal R§D support corrects these deficiencies. However, there is an economic rationale for some federal support of basic R§D that might aid automotive and other technologies. Companies cannot fully capture the benefits of long-range R§D, and therefore long- range R§D is insufficiently funded by the private sector. The criteria for additional federal support, however, should include joint private/ public sector funding, project selection and supervision, as well as full disclosure and public ownership of results. Perhaps an automotive equivalent to Electric Power Research Institute (EPRI) is needed, but it must be structured with the recognition of the fact that the automobile companies do maintain large laboratories accomplishing long-range R§D that might lead to proprietary technologies in a competitive industry, whereas utilities have not supported such facilities and proprietary technologies are not as relevant in a regulated industry. The result of the Research Safety Vehicle (RSV) program also raises some disturbing questions. DOT displayed its RSV vehicle with great fanfare only to have the president of General Motors confront a secre- tary of transportation on public television with the fact that GM's X- body car had equivalent safety characteristics, better fuel economy, and more occupant space than the RSV car. Thousands of X-body vehicles were being produced per day at the very time the RSV was unveiled. DOT personnel have confirmed the observations of GM's president. Thus the RSV program does not seem to have been a good use of federal funds. In summary, the federal government can probably do little that is helpful by direct R§D support of automotive technology. However, a great deal could be done to promote cost-effective innovation by much better federal performance in evaluating and setting automotive emis- sion and safety standards. A few specific suggestions are in order: l. The mandatory use of seat belts is an innovation that has resulted in more than 75 percent of utilization of harnesses in appro- ximately 20 countries. No U.S. secretary of transportation has attempt- ed such a program despite innumerable studies demonstrating that this is the single most beneficial safety innovation available. No foreign country has adopted the U.S. approach of passive restraints. Why is there a barrier to this innovation in the United States? 2. Carbon monoxide is the one pollutant that can be personally monitored. This can be done by measuring the carboxyhemoglobin (COHb) content of blood. The current ambient monitor of CO is a poor indica- tor—its measurement varying by factors of at least two depending on trivial changes in location. EPA has not accepted this innovative and more accurate approach to measurement. This in turn increases the stringency of emission standards because of uncertainties, and this then limits the range of automotive innovation. 3. Reducing hydrocarbon emissions in the Los Angeles basin reduces oxidant levels. Reducing hydrocarbon emissions in the Northeast has no impact on oxidant levels. EPA has refused to contemplate the l84

innovation of a regional approach to automotive emission control, even though the California standards have always differed from federal stan- dards. This barrier to innovation ripples through the system causing needless expenditures. DOT STIMULATION OF INNOVATION IN AIR TRAFFIC CONTROL--WHERE IT IS A CUSTOMER BUT NOT A USER Government involvement in air traffic control, where the government is its own customer, is and should be different from federal involvement with automotive innovation. While the users of airways are mostly in the private sector, they utilize a federally operated air traffic con- trol system and must comply with its procedures and safety and equipage regulations. Their economic fate and their physical safety depend on the safety and efficiency of the air traffic control system over which these users have little control, except by appealing to Congress or the FAA. However, the FAA has been sensitive to this problem and has insti- tuted formal user consultative conferences and interchanges to assure user input to FAA engineering and development (E§D) initiatives. Some user comments resulting from this process illustrate the value of a formal user consultation.l The user community did develop some general conclusions concerning operational restraints on E§D objectives. For example, all users recognize the need for evolutionary development of the ATC system—not as an excuse for slow development—but as a recognition of the limits to change in a system that operates in real time with many lives at stake and with massive investments in the training and proficiency of hundreds of thousands of people and measured in the tens of billions of dollars of equipment. This evolutionary requirement is certain to cause complications, expense and delays in upgrading center and terminal automation. The FAA must obtain whatever manpower and money is required to accomplish this vital program. A constant concern of the user community relates to the length of time taken to complete and implement certain vital E§D programs. For example, M§S has been under development for a decade and still has many remaining uncertainties so that an eventual implementation date is simply not in sight. The rate of development of the Vortex Avoidance System (VAS) is of equal concern. Meanwhile the airport capacity issue becomes ever more serious. The user community is also concerned about the need for improved integration of E§D programs within the E§D struc- ture of FAA, with other relevant organizations in FAA and with users and manufacturers. The troubled introduction of autoland is an example of the problem. Pilots first attempt automatic landings under better visibility conditions than the minimum certification of their equipment. This provides 1 Quotes are from "New Engineering and Development Initiatives—Policy and Technology Choices," DOT-FA77WA-400l, vol. l, March l, l979. l85

early familiarity with the equipment in a forgiving environ- ment. However, the ILS signal quality is less satisfactory in this environment than under poor visibility conditions, such as CAT II, when aircraft must avoid areas that adverse- ly affect the ILS signal quality. The autoland system follows the ILS vagaries faithfully, but the pilot is sure he can accomplish a better landing manually, so the pilot decouples, his familiarity suffers and his reluctance to use autoland increases. Unfortunately there are other inconsis- tencies between some ATC procedures and autoland capabilities. Furthermore, the reliability specification is unrealistically high. Therefore its complexity is great and maintenance expensive. When aircraft operators realize autoland is not used frequently by pilots, they are less fastidious about its maintenance. This discourages pilots even more. Obviously, coordination between pilots, manufacturers, operators, FAA flight standards and MLS advocates is needed if autoland is to become a reality as NTSB suggests. Could an organization or process within FAA coordinate all the participants in an effort to achieve utilization of autoland? Another example has to do with airport capacity. Exquisite integration is needed between runway, exit and taxiway design, terminal automation, M§S, wake vortex avoid- ance, MLS and surveillance of the surface and the terminal airspace, in order to squeeze capacity into airports safely. Could an organization or process within FAA—perhaps as an extension of the present Airport Task Forces—integrate the various components needed to improve airport capacity on a site specific basis? One last example deals with upgrading the air traffic control process in centers and terminals. This is a huge and necessary undertaking. The development of the desired ATC capabilities requires significant effort in two areas: first, the establishment of new automation concepts, the related operational procedures and the corresponding computer algorithms; second, the procurement and implementation of the necessary hardware and software to support the automation requirements. The first of these two tasks may well be the most time consuming and difficult since it involves explora- tion of some fundamental changes to the ATC process itself. The removal of rotating beacons from airport terminals and compass locators from outer markers has caused pilots unnecessary difficulties measured against the trivial cost of maintaining these facilities. In some cases, pilots will not accept a visual clearance to an airport on a clear night because they cannot identify the terminal against a background of urban lighting in the absence of a rotating beacon. This decreases aiport capacity and increases controller work load. While this issue is not as significant as most E§D policies discussed previously, it is included here to illustrate the value of formal user consultation. l86

These quotations indicate the value of formal consultation with private sector users of the government-operated ATC system to be sure that federal programs respond to perceived needs. The users know they fund E§D products through the trust fund and were not reckless with their suggestions. As one can see from this abstract of user comments, FAA has had difficulty completing programs in a timely way owing to money limitations and the long and burdensome process of obtaining approvals for technological developments through FAA, DOT, OMB, and the Congress. DOT STIMULATION OF RAILROAD INNOVATION--WHERE IT IS NEITHER CUSTOMER NOR USER The Federal Railroad Administration (FRA) also has a problem making sure that its R§D is consistent with the needs of the railroads. As was in- dicated previously, some early programs were driven by technology rather than market needs, and the results were unfortunate. In recent years, FRA and the Association of American Railroads (AAR) have developed a way of jointly funding R§D programs. An evaluation of this process for AAR concluded^: ...The AAR's policy of bringing 'funds to the table,' negotiating joint programs and maintaining a high level of R§D competence within AAR to orchestrate the effort is an excellent approach to public/private sector management of R§D. Neither the Air Transport Association (ATA) nor the American Public Transit Association (APTA) has done as well as AAR in recent years in modulating federal R§D programs to meet user needs. This is the case despite the fact that APTA represents local governments—pub lie bodies—while AAR represents mostly private sector interests. AAR's better record seems to be due to at least two factors; neither APTA nor ATA invests money in R§D and neither organization has a strong R§D capability. As a result, members of ATA have had to install safety equipment of dubious value developed and mandated by the FAA; and APTA has had to endure various bus and train demonstration programs that they felt were inappropriate and badly executed and that have led to mandated equipment they do not wish to operate. DOT STIMULATION OF TRANSIT INNOVATION--WHERE IT IS AN INDIRECT CUSTOMER BUT NOT A USER The Urban Mass Transportation Administration (UMTA) funds R§D that it hopes will be useful to the transit properties, most of whom are supported by 2 "Joint Public/Private Sector Management of Railroad R§D," Economics and Science Planning, Inc., May 24, l977. l87

capital and operating grants from UMTA. Many issues face: ...federally supported R§D when (l) the ultimate customer is not the federal government but local transit properties; (2) the ultimate user is the public with many diverse interests; (3) the manufacturers' investment in development and tooling is predicated not only on market forces but also on federal and local regulations and procurement policies and most particularly their steadiness; and (4) there are government concerns about anticompetitive concentration. Some observers have described transit system develop- ment under these circumstances as follows: The user is not the buyer, the buyer doesn't pay for it, the payer usually doesn't buy it or use it, the operator is a professional who doesn't use it, pay for it, or buy it. An Office of Technology Assessment report4 evaluated some early UMTA efforts as follows: The Transbus program focused on a product rather than development of the key components that would make that pro- duct practical in revenue service. Fearing increased main- tenance and reliability costs and service penalties because of the uncertain status of key components, such as canti- levered tires, brakes and axles, the transit operators with- drew their support. Transbus as initially proposed and con- ducted was overly ambitious, expensive for what was accom- plished and delayed an interim or advanced bus. Future financial support for Transbus could be used more effective- ly if directed at component development and evaluation. The transit operators view the SOAC (State-of-the-Art- Car) demonstration with mixed feelings. To some it was use- ful, to others it was not. Urban Mass Transportation Ad- ministration (UMTA) during the period l968-l972 took the approach that aerospace technology and management techniques could offer substantial benefits to the transit industry. In this spirit, UMTA determined that a SOAC demonstration would be useful even though it incorporated no new technology. The ACT (Advanced Concept Train) program was overly ambitious, troubled with unrealistic cost estimates, late deliveries and management problems. ACT incorporated in its subsystems several important technological innovations which, when proven, are likely to be adopted by transit properties. An ACT program aimed at subsystem development and evaluation rather than construction of an integrated vehicle might well have been more effective at less cost. Present UMTA manage- ment has recognized this situation and has instituted an Advanced Subsystem Development Program with these objectives. •* "Transit Vehicle R§D--Transbus, SOAC and ACT," United States Congress, Office of Technology Assessment, March 8, l977. 4 Ibid. l88

UMTA should support research, development and demonstra- tions of train and bus advanced technologies, particularly at the subsystem level, that would lead to improved transit vehicles. It is not an effective use of federal money to develop conventional transit vehicles as final products, since transit manufacturers can more effectively do this based on specifications developed by the transit operators. These specifications will incorporate new subsystems appropriate to each local transit property when they are shown to be effec- tive by UMTA or industry. Standardization at the subsystem level is frequently achieved through a procurement process that permits transit properties some discretion. Most of the drive trains and engines on current buses are standard without federal regula- tion other than the consent decree of l965. UMTA's heavy rail car standardization efforts at the subsystem performance and interface level promise to be beneficial. Standardiza- tion of the total vehicle design is likely to be unproduc- tive. The ability to deliver R§D to the transit properties is dependent on federal policies relating to procurements, grants, regulations and standards as much as on the R§D pro- cess itself. These interactions and some alternatives are discussed in the report. This assessment has focused on conventional transit vehicles. Other policies may be appro- priate to high risk advanced transit system development. There has also been concern with advanced transit system develop- ment. One expression of this was the formation in l976 of the Advanced Transit Association (ATRA)—a group of urban planners, transit techno- logists, and transit operators who felt that APTA represented transit system operators but not necessarily individuals who used transit sys- tems or urban planners. Its purpose is, "(l) To improve the quality of urban life through the judicious application of advanced technology and planning concepts to transit services; (2) To disseminate informa- tion on advanced transit to the members, to the interested professions, to the public, and to representatives of all levels of government; and (3) To improve the quality of transit-system analysis, planning, design and implementation." Another expression of concern has resulted from a recent review of European, particularly German, progress in advanced technology transit, occasioned by the International Transportation Exposition in Hamburg in June l979. The consensus of congressional, DOT, and private sector individuals is that the United States is substantially behind demonstra- ted German technology whether it be M-Bahn (magnetic levitation), C-Bahn (CRT or PRT), S-Bahn (commuter rail), U-Bahn (rail rapid transit), or Strassen-Bahn (street rail). One irony in this situation is that the two outstanding texts on high-technology transit were published recently (l978) by two Americans—Transit Systems Theory by J. Edward Anderson and Fundamentals of Personal Rapid Transit by Jack H. Irving. There is nothing wrong with the U.S. ability to conceptualize and analyze, but l89

evidently the United States does not produce prototype hardware. A U.S. urban planner has to travel to Germany today to view the full range of transit options. DOT lack of support for developing and testing tran- sit options has led to this debacle. Some quotations from the trip reports of several federal U.S. observers make the point: Volkswagen, BMW, Mercedes, Siemens, MBB, and many others had exhibits featuring new technology in electronic controls and safety systems. The most disappointing exhibit was the U.S. exhibit put on by DOT. It featured the minicars RSV, the UMTA paratransit vehicle, an air bag display, and photo- graphs of President Carter and Brock Adams. On Sunday I flew to Hamburg to attend and participate in the IVA. A great deal of money and effort went into the exhibits themselves with representation primarily from European countries. The European supply industry displays of equipment and new technology developments were impressive. To the contrary, the U.S. exhibit was somewhat embarrassing for its lack of any significant equipment or technology offerings. I diagnosed with some chagrin and envy that...it is now a virtual certainty that at least two Japanese systems (more ambitious than ours), one French system, and one German system will become operational before we open our first DPM.... To me, it seems that other industrialized nations can move much faster in making decisions and implementing them than the United States....We have a long way to go to obtain this kind of commitment by transit to consider new technologies. ...a feeling of a truly cooperative joint effort between government, transit operators, and the equipment suppliers in developing new transit technologies and techniques, is needed as well as more cooperation than I believe exists in the United States today, although it is possible that the harmony was more a promotion for the visiting delegation than in reality exists. Still, it is difficult to imagine an experience such as the Transbus controversy taking place in Germany.... Thus UMTA has evidently failed to stimulate innovation where other countries have succeeded. CONCLUSIONS AND RECOMMENDATIONS It is clear that transportation innovation has not prospered under DOT management. During the first half decade of DOT's history (l967-l972), technological developments supported by DOT too frequently lacked a market. Evidently, there has now been an overcorrection—at least in certain modes—and technological possibilities and even requirements are not being developed at the needed pace or in some cases at all. l90

The modal administrations within DOT seem to have learned to work with users in recent years so that technological developments tend to satisfy user needs. But the technologists within the modal administra- tions have not been able to push through the thicket of bureaucratic stifling of innovation within their own modes, within the DOT super- visory structure, within OMB, and within the Congress. Maybe no mortal could. Perhaps the bureaucratic maze should be unravelled. One example should suffice: In Germany, the Ministry of Research and Technology has a staff of two devoted to urban transit development. UMTA has a staff of 65, many of whom are evidently consumed in modal, DOT, OMB, and congressional justifications and contractual procedures. Recall that with this staff of two, Germany is outperforming the United States. Recommendation: Streamline the technological approval process in DOT and the federal government. There is no political stability in DOT. There have been five secretaries of transportation in l0 years, as well as five UMTA admini- strators. Not one of these rose through the ranks, a sign of a badly managed enterprise. Each secretary tended to reverse the policies of his predecessor. Volpe wanted to mandate air bags, Brinegar felt the decision should be left to the private sector, Coleman wanted to test air bags, Adams mandated air bags, and now we have Goldschmidt. Volpe wanted to mandate Transbus, Brinegar wanted to leave it to the transit operators, Coleman selected Transbus parameters acceptable to operators and manufacturers, Adams mandated more stringent parameters, and opera- tors and manufacturers "got off the bus." Innovation cannot prosper in such a highly politicized department. Recommendation: Depoliticize DOT, at the very least depoliticize the technological components of DOT. Consider whether modal administra- tors could be bipartisan, appointed for six-year terms, preferably from the ranks. A change in administration in the United States involves a change in 4,000 top government professionals, usually within six months. This is an impossible task to do well. Germany, England, France, and Japan change only 40-70 top jobs with a change in administration. They seem to maintain political control. Innovation seems to do better with federal stimulus where the federal government is the customer—air traffic control—and least well where the government is neither the user nor the customer--the automo- tive industry. Recommendation: Deploy R§D assets where it is more appropriate for federal involvement in technological development, that is from the auto- motive industry to air traffic control, for example. Railroad and transit R§D are also more appropriate recipients of federal technologi- cal development aid, since the long history of federal regulation in railroad transportation stifled innovation and adaptability and since the federal government funds most transit capital gains. l9l

DISCUSSANT'S COMMENTS BY HOWARD K. NASON There has been general agreement at this meeting that technological inno- vation comprises the successful introduction and diffusion of new pro- ducts, processes, or services, using new technologies or new combina- tions of technologies that distinguish the innovation from its predeces- sors. There also has been a consensus that, in the United States at least, the creation and introduction of such innovations is a function of the industrial sector. But the essential final step of the innova- tion process, commercial acceptance, is controlled by the public. Inno- vation has not been accomplished until the user makes the critical deci- sion to buy it. Morlok and Goldmuntz, of this panel, and Garrison, of the Panel on the Setting for Innovation, examine in their position papers many of the complexities of the innovation process. They specifically probe the many factors involved in innovation in transportation. From their contributions and from the views of other participants in the workshop, we can see a number of areas of common concern. THE ROLE OF TECHNOLOGY R§D and the technology it generates are essential elements of innova- tion. Standing as they do at the beginning of the process, they have been taken by many to be the most important element in the chain. Speakers at this meeting separately have emphasized that every link in the chain is essential and that if any one is defective the entire pro- cess fails. Capital to carry the technology into production; marketing skills to insure a match of user needs and wants with ability to manu- facture and to secure consumer acceptance; and overall management to insure integration of each step in the process, to make sure that social as well as economic and market requirements are served by the innova- tion, and to secure a return sufficient to repay the costs of the inno- vation and to help support development of successive ones, are all required. Thus a climate supportive of all steps in the innovation process is essential. It is the role of management, whether in industry, govern- ment, academia, or other institutions to insure the perpetuation of such a climate. Technology thus is a concern of all elements of our society. There seems to be agreement that technology per se is in good health in America, with support in all sectors. Specifically, there seems to be no lack of innovative components and processes for application in l92

transportation. There does seem to be a shortfall in the understand- ing of total systems, into which components and processes must be integrated if they are to be useful. Also, knowledge of valid user needs and wants, as they relate to various modes or combinations of modes, and to the opportunities for innovation in these, seems to be in need of improvement. Both private and public sectors need to develop better capabilities in these areas. Charpie cautions that user needs are not always what users say they want. PUBLIC ATTITUDES Public opinion is important to innovation, not only with respect to acceptance but also with respect to the support of R§D, to the genera- tion of investment capital, and to the maintenance of a favorable climate for constructive change. Several speakers expressed concern over the pejorative antipathy between public and private sectors that is evident today, intensified by misguided utterances that seem to receive more attention from the media than is warranted by validity or merit. Public ignorance of elementary economic realities is especially damaging to innovation. Lack of understanding of investment, profit, savings, and of who ultimately must pay the costs, leads to political pressures resulting in actions very damaging to savings, creation of capital, investment, and modernization, which are essential to innova- tion, productivity, international competitiveness, and the maintenance of quality of life. It is felt that government should take a positive, rather than an adversary, approach to such problems. ROLE OF GOVERNMENT On a number of issues there seems to be a consensus, both in the pre- pared papers and in the discussions. Such issues include the following: l. Support of basic knowledge. While industry and universities will continue to make substantial contributions to the support of basic research, the majority of the funding should continue to come from the federal government. (This holds true for most other industrial nations also.) Most of the actual performance of basic research should continue to be done in the universities. 2. Support of generic technology. Where the development of basic knowledge into technology that is generic to an entire industry or to several industries, and where the cost of such development is too great for any company or group of companies to bear, support by government is appropriate, subject to reservations discussed below. Such support is most effective when industry and academe participate in planning, fund- ing, and execution. 3. Establishment of objectives. Government could make a major contribution by catalyzing the establishment of overall, long-range objectives for innovation in a technology or in an area of social need. l93

Nowhere .is this more evident than in transportation where national needs should be evaluated from the standpoint of total systems, of their in- teractions, and of their relationships to other national goals. Defini- tion of needs for innovative technologies would flow from such efforts. 4. Climate. Government plays a critical role in determining the climate for innovation and productivity. It should make sure that its actions contribute to a climate favoring initiative in the creation and application of technology to the total innovation process, and parti- cularly should emphasize the avoidance or the removal of disincentives that its other activities may create. Control of inflation is an essential step. Elimination of confiscatory tax policies, and the adop- tion of policies favoring savings and investment are needed to fuel vital steps of the innovation process. Regulation and direct control of technical and economic matters must be handled so that innovation is encouraged rather than inhibited. And management processes in government must be depoliticized if they are to be effective. 5. Direct intervention. Government should intervene directly in the latter stages of innovation only where market forces clearly are incapable of meeting a national need. Examples include streets and highways, waterways, public health, area sanitation, etc. Public acceptance, as evidenced by willingness to buy a new product or service, remains the most potent decision-making element. Government should avoid intervening in product or process decisions and should guard against attempts to impose new technologies, however otherwise attrac- tive they may appear to be. ROLE OF THE PRIVATE SECTOR The private sector has responsibilities to insure diversity in the creation and delivery of goods and services that meet society's needs. The public is the ultimate judge of success; it buys or it does not. Not every innovation—not even most innovations—will win such accept- ance. An industry's overall track record in winning such acceptance will determine its survival. Not every innovator will survive. Many will not deserve to survive. That some deserving ones will not survive is unfortunate, but survival of the fittest is the determinate of fitness. Artificial props to prevent failure of unacceptable innova- tions never succeed in the long run. Industry thus has fundamental responsibilities, to the public and the consumer, to its employees, and to its investors. Failure to meet any of these responsibilities fully constitutes failure of the whole. INSTITUTIONALIZATION Transportation is highly institutionalized between transport companies themselves (railways, airlines, shipping companies, bus and truck companies, etc.), regional operators (e.g., transport authorities, Amtrak, Conrail, etc.), suppliers (component and equipment manufacturers), l94

associations, and a variety of governmental bodies involved in every phase of the process. This affords an opportunity as well as a challenge. Opportunity for an intermodal engineering systems approach, in which all sectors, public and private, could participate constructively, from earliest conceptual and planning phases to final delivery and operational phases. How can we integrate (compromise) public and private roles in decision making? We have historic models that provide clues as to how or how not to do it. One of the best of these is the example of the former National Advisory Committee on Aeronautics. NACA never designed airplanes. But it provided basic technology of great sophistication, which the industry then incorporated into advanced designs, which led to aeronautical pre- dominance for America. Industry, academia, and government working together produced superior basic technology, and through its applica- tion, innovation. NASA, which absorbed NACA, on the other hand, has lost the touch for this kind of collaboration, and innovation has suffered. Examples from the Department of Defense as contrasted to those from the AEC/ERDA/DOE provide the same kind of lesson. Bob Charpie drew stimulating conclusions. In sum, those in attendance came through strongly for a participat- ing systems approach for innovation in transportation, with appropriate inputs by government, industry, and the academic community, not domina- ted by one sector, but guided by the fundamental principles that have been shown as controlling for the process of innovation. A readjust- ment of the role of government, and of its image of that role, clearly is indicated. l95