Let us begin this concluding chapter by reviewing some commonplace but fundamental ideas about R and D. As a framework for organizing the evidence presented in the previous chapters into conclusions about the peculiarities and effectiveness of Soviet energy R and D, it will be helpful to restate briefly what R and D is and what some of the major isssues in R and D management are.

Research and development is a process of generating new knowledge and applying it to solve production problems. As this definition suggests, the process encompasses a number of quite different activities, from basic research to the commercial introduction of developed technologies, commonly thought of as lying along a spectrum of decreasing uncertainty, greater applicability to production concerns, and increasingly heavy resource commitments. The search for new knowledge that constitutes one pole of the spectrum must proceed in a way that is not specifically focussed on the task of solving a recognized production problem. At the other, actual introduction of new technologies that raise productivity must take as their point of departure, capabilities and potentials that are known with a considerable degree of assurance. In between is a great variety of studies, experiments, and evaluations, characterized by intermediate degrees of relevance, uncertainty, and expense. One of the universal tasks in any R and D enterprise is the balancing of effort among the different activities along this spectrum—between long-range speculative efforts intended to create quite novel solutions versus efforts to make more modest advances on the basis of fairly secure expectations of what will work. In bulk power transmission, to take a concrete example, there is a question of whether to seek increases in the amounts and distances involved, through small incremental changes in the parameters of a familiar approach or to abandon that as inadequate to the task and explore instead such novel concepts as cryogenic conductors or wave guides.

This balancing is not, however, primarily a problem of choosing one or the other option or even of static allocation between the various parts of the spectrum. Rather, the division of efforts along the spectrum at any given time should grow out of a dynamic process of deciding when it is time in any given area to advance on the basis of the new knowledge built up by speculative explorations to a narrower and more focused exploration of possible solutions to some problem. When has enough knowledge been gained about the behavior of a certain kind of coal to justify taking the concept of energy-technological processing from laboratory experiment to a pilot installation, from a pilot installation to an experimental plant, and from an experimental plant to a demonstration or commercial facility? When has the environment surrounding a problem so changed as to make a research idea or direction irrelevant, so that it should be dropped and the effort shifted to an alternative lactic or back to rethinking the fundamentals of the phenomena involved? When has enough experience been acquired so that it is possible to decide on a big resource commitment with a very high probability that the specific solution proposed can be made to work in a situation that will now involve the acquiescence and cooperation of a much wider circle of actors?

At this commercialization stage someone will have to design and produce the associated equipment to meet performance specifications, while others will have to be willing to accept it, fit it into their environment of production goals and conditions, and make it work in practice. Success here is a function not only of how well all the interrelated tasks at this stage are carried out, but also of how well previous steps have set the stage for this final commitment. For example, the preparatory decisions must not have mistakenly neglected better alternatives that will subsequently emerge to undercut the effort put into this one—if fission reactors are more advantageous than breeders, even the most dogged attempts to commercialize the breeder will be a fiasco. The developers must have performed enough work on components, materials, and production processes to assure that the new equipment, facility, or system can be made to work; they must have adapted the final outlines of the technology well enough to the conditions that will be faced so that it will be accepted as economic. Only then can it succeed in being widely diffused throughout the area of application.

Against the background of this description of the process of R and D, what is distinctive about Soviet R and D experiences and approaches? What kind of systemic peculiarities seem to account for the results we see?

I find it difficult to summarize the energy sector R and D activity we have described in the form of generalizations that hold up in all cases, but there does seem to be a number of distinctive features common to most of the examples we have looked at. At the same time there are important differences from the energy R and D experience of other countries, and one of the goals of this final summary should be to reflect not only on the way Soviet system characteristics make R and D outcomes different from what we see elsewhere, but also to seek the variables that condition differences of outcome among cases within the Soviet system itself.

First it is perhaps obvious, but worth repeating, that in many of the areas we have examined, Soviet technology planners were following development paths already well explored by other countries. In energy as well as in technology generally, the Soviet Union has been and, in many areas is still, on a lower technological level than the advanced market economies. The experience of other countries thus offers helpful guidance in deciding the best directions to follow in technological advance and in dealing with the basic difficulty in all innovation—eliminating uncertainty as to whether some idea can be made to work. The shift to narrow-web combines, the scaling up of excavators, use of trucks in strip mining and continued increase in their capacities, the use of gas turbine equipment for peak power generation are only a small sample of cases where a technological line had been chosen, developed, and demonstrated to be effective in practice elsewhere well in advance of Soviet efforts. The Soviet planners could both take that experience as indicating the correctness of the line and operate with the assurance that the new technology could be made to work.

The fact that in many cases R and D programs represented domestic catch-up along well-established paths has had an important impact on the R and D process by encouraging the location of innovation decision making at a high level in the hierarchical structure. There was a presumption that experts at the top, possessed of a special vision, could best discern the key technological trends in any area and choose the most appropriate ones to follow. This kind of centralization gives a strong strategic flavor to the whole process, which encourages both entrenched biases at the center that may be difficult to overturn through experience at the bottom and selective attention that may leave important tasks neglected. Recall for example the negligence toward developing peaking equipment compared to the tremendous stress laid on mastering large-size generating units and stations.

This study has been interesting to me in showing that in many cases the technological inspiration has come from countries other than the United States. Long-wall mining techniques seem to owe nothing to the U.S. example; the experiment with tidal power has drawn on French experience and even French equipment; the Soviet strategies of nuclear power development have rejected the U.S. line on breeders and have accepted the French and British view that they are feasible and economically attractive. There are even a few cases in which the source of the inspiration was Eastern Europe, notably in the case of rotary excavators. It is a little surprising to me that there has not been more borrowing from Eastern Europe. I would think, for instance, that Polish coal-mining technology would offer a number of examples of technological improvements the Soviet Union could emulate, especially since Polish mines work in a similar system setting.

When technological progress takes place in this manner, one of the corollaries is that R and D is less a matter of picking a way through the unknown than of choosing from a menu of demonstrated possibilities. The problem is to determine which lines of advance already pioneered elsewhere are economically appropriate under Soviet conditions and to assess how ambitious an advance domestic R and D performers can be expected to manage. I don’t believe this study has shown that there is a uniform conclusion as to how well this job is done. On the whole it is my impression that Soviet energy technology planners do have a healthy sense of the peculiarities of their environment and are reasonably cautious in deciding what new ideas are feasible for them to try to develop domestically. The great emphasis put on the heat rate as a design objective for power generating equipment is a natural response to the relatively high cost of fuel in the USSR. I have been impressed with what seem to be rather reasonable approaches to the economic analysis of alternative technologies in electric power generation, with the ideas underlying the fuel optimization models, and with the approach to the economic evaluation of nuclear power. Using larger pipe diameters to compensate for deficiencies in attainable working pressure and compressor equipment should probably be judged a reasonable accommodation to the difficulties the pipeline designers see in trying to match Western steel and compressor qualities.

We have seen a few cases in which decision makers were overoptimistic about the ability of domestic innovators to solve some problem, resulting in the failure of R and D programs to achieve their objective. This would be my interpretation of the problem with gas turbines, for example. Very often, however, this kind of overoptimism is not fatal. There is often some way to deal with failures to meet technological objectives, as when the electric power industry compensated for the lower quality of steel by accepting operating temperatures for power blocks below those the engineers thought they had designed for.

Despite this generally emulative approach in energy R and D, however, we have shown that the system is perfectly capable of supporting work in novel areas and at the frontier of technological advance, even when the direction chosen has not been validated by foreign experience. Ideas and possibilities that are a long way from realization because they involve basic research or because their potential must be evaluated through a long period of experimentation do get supported. The Soviet energy “establishment” itself has been thoroughly unenthusiastic about either solar or geothermal power, but research programs in those areas have gone on through the influence of alternative sponsors such as the Academies of Science. Even organizations that might be expected to be biased toward current production tasks have often supported long-shot ideas—Minenergo has been willing to spend resources on tidal power and to risk prestige and quite large resources on MHD. It is interesting to try to understand how such work is protected from the short time horizon that indeed dominates most decision making in the USSR, as well as from the pressures to focus on solving immediate production problems. One explanation is that, despite operating with a generally mission-oriented perspective, R and D planners understand well enough that there should be a distinct mission conceived of as building up a scientific-technical backlog of ideas that might be useful at some future time—the so-called nauchno-tekhnicheskii zadel. This is reinforced in execution by a kind of institutional decoupling—the designers of the R and D system really do try to set up organizations that will be independent of the production people with their focus on current problems because they have high-level protection and independent financing. This is the positive side of the phenomenon noted in Chapter 2—the inertia and insensitivity of many R and D organizations that comes from assigned specializations and the prevalence of institutional, rather than project, financing.

But I believe we must also accept the idea that the high-level decision makers in the production establishment, who in the Soviet setting do play a dominant role in the system, are capable of vision, and are willing to accept significant risks in supporting R and D. There are examples of research directions which have had continued and substantial support even though they may be at variance with the general directions of world technological development or have been consciously rejected by decision makers in other countries. Support for the MHD program has continued through the ups and downs this idea has suffered in other countries. Similarly, within an example not discussed in this book—the turbodrill—the oil industry supported through thick and thin a completely distinct line for improving drilling technology (see Campbell, 1968). In both these cases I believe the explanation must be partly in terms of personal and bureaucratic influence. When a bureaucratic structure or a powerful individual is really convinced and committed to some direction, it will be doggedly pursued despite discouragements or the contrary example of other countries. As another example of independent innovation, the Russians have committed themselves to cogeneration essentially on the basis of the energy conservation argument, even though this was neither strongly born out by their own experience or validated by the practice of other countries. In this case, the explanation is partly a difference in the institutional setting—central planning means that the heat and power markets were subject to unified control, and the technological decision makers had the sense to exploit the novel opportunity thus presented.

I would like to suggest that the various devices that protect basic and speculative research in the system exact a serious penalty by complicating the problem of moving smoothly from one stage of the science-production cycle to the next. This is one of the most serious weaknesses of Soviet R and D and will be discussed more fully below. Here I want only to note that the institutional decoupling that protects speculative and basic research from the attentions of officials with short time horizons and production orientations also makes it difficult to accomplish a smooth transition from one stage of research to the next. Everything is either rather speculative research or overhurried movement toward industrial application. We saw this in the case of MHD, in the development of big power blocks, and in the nuclear program; I will expand on this “commercialization” problem below.

This study of energy experience suggests that Soviet R and D institutions probably fail to generate at each stage enough alternatives for evaluation, a finding that corroborates a widely-accepted generalization in the Western literature on Soviet R and D. In Chapter 6, it was shown that importation of technology as an alternative to domestic innovation seems not to be a routine option in the economic analysis of R and D decisions. There is a compartmentalization of the decisionmaking environment which requires some extraeconomic bureaucratic decision for technology transfer to be given serious consideration. There is not enough institutionalized competition to generate alternatives at the middle and final stages of choosing technologies. A number of cases of competitive efforts in development have been mentioned— the design competition for drilling machinery for strip mining, the VVER versus the RBMK nuclear power reactors, alternative types of compressor equipment for gas pipelines (which incidentally means recourse to different plants and ministries), alternative designs by two different plants for the 500 MW power generating block. Nevertheless, in many of the cases examined there is a strong element of monopoly, in the form of an established set of ideas, and a shortage of institutional and program alternatives. Large trucks are a case in point, and the dependence of the coal industry on machinery plants uninterested in its problems is another. There are four plants producing excavators, but they have deliberately been given a strong intraproduct specialization. The fact that the line of excavators developed at one of the plants was actually assigned by the Ministry to another plant to be produced is an interesting reminder that the customer is really facing one high-level decision maker, not two potential suppliers. This limitation of alternatives is an unsurprising manifestation of the seller’s market characteristic of the Soviet-type economies. One of the most interesting things to me is that Minenergo as one big buyer seems to have been more successful than other industries in getting what it wants in the way of technological advances. Minenergo is perhaps the closest analogue in the civilian sector to the Ministry of Defense. For some producers it is essentially the only customer and hence in an advantageous position to exercise control over their innovation efforts and to exert pressure on their performance regarding quality—advantages commonly thought to be important in the Ministry of Defense’s generally successful record of technological advance.

At various points in the discussion I have touched on the question of where among the various participants in the R and D process the weakest link is. When we find an example where the technological level seems to be low or technological upgrading proceeds slowly (inadequate equipment for open-pit coal mining, poor reliability or efficiency of gas turbine equipment, technologies poorly oriented to the needs they are supposed to serve), can we point to the performance of any particular class of participants as being the major bottleneck? There is some tendency in the literature on Soviet technology to say that the weakest link in the cycle is at the development end, especially in the behavior of the firms who are supposed to produce equipment embodying new technology.

There is perhaps some support in this study for that idea. Apparently in most of our cases the people responsible for basic research, for making technical and economic evaluations of alternative possibilities, and for designing new technological systems, are able and well trained. In a number of cases we would have to acknowledge them as bold and inspired. The work on fusion, on MHD, on high-voltage power transmission are cases in point. The testimony of coal industry officials is generally favorable as regards the designers, who have come up with concepts and designs adequate to solve the problems the producers have identified. And certainly in a number of cases we have looked at (especially in the coal industry), it is the ineptitude or disinterest of the equipment producers that has caused the most serious delays and disappointments in upgrading the technical level of the industry. But I think that it is fundamentally untenable to identify this as the weak link. In most of the cases presented here, there is a compound failure of responsibility involving many of the actors and a number of the stages of development.

The people who design open-pit mining schemes themselves contributed a prejudice that probably delayed efforts to produce excavator and truck models adapted to the needs of the mines they were operating. The designers of the gas-steam combined cycle prototype made an error in assuming that gas or low-sulfur fuel oil would be available, with the result that the development work done to date on that idea represents, to some extent, time lost on a sidetrack. But I doubt that we can really fault them for such a lack of foresight—they seem to have done their work well within the constraints of the problem as it was assigned to them. Many of the cases suggest that the system pushes the designers and manufacturers into premature efforts on a timetable that does not allow enough preliminary work—as in the decisions to produce 500 MW and 800 MW generating blocks when experience or information necessary for designing boilers that would work on the intended fuels was simply not available. (It is impossible to know to what extent the R and D performers themselves were remiss in not insisting on a timetable that would permit them to do their job right.) Often, the complaint about equipment once it is produced is that it doesn’t really fit the conditions encountered, but it sometimes turns out that these conditions were never properly communicated to the designers.

Innovators get caught in a web of interdependencies that frustrate what may well be honest and intelligent efforts to raise the technological level of their outputs or operations. I believe the coal industry planners really have come to accept that trucks and larger excavators could improve the economic conditions of strip mining, but their aspirations in this direction are now constrained by the actions of the truck and excavator producers. The excavator producers in turn say that they cannot obtain steel of the appropriate qualities to meet the specifications the coal industry has articulated. This breakdown through interdependence underlines a thesis that the weaknesses in Soviet innovation probably have less to do with technological knowledge proper than with organization. Of course this is not inconsistent with the way innovation is usually understood—central to Schumpeter’s concept is the fact that the innovator has to overcome problems of finance, marketing, labor, and so on, as well as purely technical problems. But one does develop a feeling that these problems are more intractable in the Soviet environment than in a more pluralistic system.

Fundamentally the problem seems to be with the weakness of feedback mechanisms in the Soviet system. We should not expect innovation to work perfectly smoothly in any economy—the penalty of working with the unknown is that some of the expectations that have guided the creators of new technology will be falsified, thus invalidating some earlier decisions and requiring their abandonment or modification. The important thing is that there be mechanisms to generate the kind of adjustments that will adapt original decisions to give second-best solutions. When one arrives at the later stages of the cycle, one is stuck with a framework of settled issues, maybe even design and production commitments, and there is no choice but to keep experimenting until the innovation is made to work one way or another. The peculiar feature of the Soviet system is that this learning often fails to induce the modifications that are so painfully discovered. Dienes says that, despite extended difficulties in making boilers work on Kansk-Achinsk lignites and the resort to extensive modification to make them work, the boiler producers just kept on producing the boilers according to the original designs. Another interesting variation on the same problem has to do with boiler-turbine disproportions. Boilers have been designed to produce steam outputs adequate to match the capacities of turbines on certain expectations about the heat content of fuels. But these expectations about heat content are almost universally unmet, leading to a wasteful underutilization of turbine and generator capacity. Under the existing institutional and incentive system, however, this kind of informational correction is not transmitted back to the attention of the designers. The explanation for all these breakdowns is the well-known failure of a strongly hierarchical structure to provide enough lateral communication or powerful enough leverage to clients to make the interactions between all the participants effective.

Rather than trying to apportion blame among the phases and the participating actors, though, it may be more useful to review our evidence from the point of view of the task of commercialization of technologies and to consider how the Soviet approach to performing this function differs from that in the United States—there are some striking differences. Furthermore, I believe that the different degrees of success the USSR has had in various technical areas are to be sought primarily in the commercialization phase and are related to the differences in the kinds of tasks posed by commercialization in different technical areas.

The distinctive features of Soviet commercialization, as I see them, might be described as follows. A decision to move ahead on introducing a significant innovation is likely to be made only when there is some unusual pressure—an outside example, a crucial bottleneck, a crash program for some sector that creates a specific pressing technological need. In the absence of such pressure, R and D work in any area of technology is likely to mark time, with little effort put into intermediate kinds of applied research, such as accumulation of data or research on materials and components, that would provide a foundation for a subsequent decision to commercialize. Once a commitment is made, however, it is likely to generate a crash program. In this effort stages are telescoped, and attention is given primarily to some components of the system involved, with little attempt to design the whole technology as an optimized system. A fairly ambitious step upward is likely to be attempted in choosing the scale of the demonstration plant, unit size of equipment, or technical parameters. A corollary of this approach is a strong preference for going directly to an expensive prototype or demonstration facility as a substitute for intensive work on the elements of the new technology to enhance predictability in design. Rather, the testing of the new technologies is likely to take place using what are intended essentially as commercial designs, to which the producers have made heavy commitments, or in facilities that are expensive enough that they must be considered for commercial exploitation. Such an overambitious and premature commitment may well result in large wastes and delays.

This description probably sounds as if I had taken the MHD program, about which we happen to know a great deal, and generalized its main features, asserting that it applies to energy R and D in general. But some reflection on the various cases considered shows that it fits quite well the development effort for gas turbine compressors, the creation of each successive generation of power generating equipment, and the experience with excavators and trucks, to mention only the most obvious examples. The report of the U.S. nuclear power reactor delegation that visited the USSR in 1974, underlines the preference for “complex or integrated reactor experiments for test purposes, versus the U.S. preference for single purpose tests,” and further elaborates:

Apparently major plant construction was started without complete designs in hand and without extensive prior proof-testing of all features. . . . It is difficult to know if the examples we saw are merely making the best of earlier oversights, but the Soviets clearly accept risks associated with the price of moving forward with demonstration plant construction. [ERDA, 1974, p. 3]

Coal conversion reveals a long history of rather poorly guided effort, followed by the scheduling of an overambitious demonstration effort. The test data and economic evaluations of the processes were based on units handling about 15–20 tons of coal a day. The next two steps, however, are to be a plant of one million tons annual capacity and then one of 24 million tons annual capacity. (As an indication of how large a jump this represents, the largest demonstration plant for coal processing now scheduled in the U.S. approach is a 219 thousand tons per year liquification plant.) Because a step of that magnitude was probably premature, Minenergo has subsequently vacillated in pressing ahead with it, as mentioned in Chapter 6. Slurry pipelines, geothermal and solar power exemplify technologies that have not yet been given the nod, and R and D in those areas pokes along with little indication of exploratory work appropriate to their current status and prospective importance and designed to provide a basis for future decisions about application.

The kind of aimless, misdirected effort that can take place is suggested by the following critique of a coal industry institute’s program to utilize coal dust to make smokeless lump fuel. This is presented in the feuilleton style of Soviet investigative journalism and may be a one-sided, but hardly fanciful, account.

Since this use of coal dust was an alluring possibility, the scientists easily obtained money to build a pilot plant.

The most logical location would have been either the Donets Basin or the Kuznetsk Basin. Instead, reasoning that it would be much more pleasant to work amidst the beautiful landscapes of the Northern Caucasus, the scientists decided to build the plant in the village of Kumysh, Karachai-Cherkess Autonomous Province. For developmental purposes they would use coal dust from the mines of Stavropolugol and then later convert to local coals when building full-scale production facilities in the Donets Basin and Kuznetsk Basin.

Time passed, and the newly built plant opened with a flourish of publicity. Since changes had occurred in the institute’s staff in the meantime, with some individuals retiring and others taking jobs elsewhere, there were heated discussions over who could perform research. At the Kumysh plant everything remain unchanged: There was smoke, but no smokeless lump fuel.

Representatives of the Karachai-Cherkess People’s Control Committee visited the ill-fated plant and saw what the scientists’ overeagerness and fantasy-seeking had led to. In their blind haste, they had accepted the production complex before it was finished. The equipment was poorly engineered and inadequate to its intended purpose.

The USSR People’s Control Committee investigated next. Again no real progress was made. Instead the institute devised a sham to give the appearance of progress. Abandoning the smokeless lumps, the researchers switched to a new project. They acquired a large shipment of coal from the Kansk-Achinsk Basin and set out to find a way to reduce its characteristic high moisture content. They drove their equipment at full blast until it was worn out. The plant had to completely shut down and major repairs were required.

A technical conference chaired by A. Mazin, Russian Republic Minister of the Fuel Industry, discussed the smokeless lump project and arrived at the conclusion that the project had failed.

It cost the state 1.134 million rubles to build the pilot plant. Five years of operation cost another 911,000 rubles. Now the pilot plant is being converted to a coal-briquetting facility so that this absolutely useless enterprise will at least be kept busy doing something. Naturally the necessary modifications require more investments. [Current Digest of the Soviet Press, vol. XXVIII, No. 51, p. 12]

The characterization I have offered may seem to differ in some respects from commonly held ideas, and we should note that it also contains an internal contradiction. A recent study of military R and D by the RAND Corporation concludes that Soviet design tends to be conservative, preferring small incremental changes rather than creation of whole new systems intended to achieve large upward steps in performance. Indeed, our characterization suggests that the ambitiousness of the step upward may be offset by conscious avoidance of developing the new generation as a complete system. To insure against failure, the energy equipment programs we have looked at tend to use as auxiliary equipment models that have already been extensively tested in practice. The innovative elements are usually introduced into a fairly conventional environment, and in this way the R and D risk can be limited to the novel elements. Recall that in the U-25 MHD plant, everything outside the MHD element itself is quite conventional. The aviation-type compressors for gas line use were based on an existing aviation engine, and the task of designing a new model from scratch was rejected. The heavy emphasis on parts commonality evidences the same philosophy in the sphere of individual items of equipment. The 300 MW power block was reported to have 65 percent parts commonality with early models, and we find the same emphasis on parts commonality in the excavator field and in the development of underground mining machinery. This form of incrementalism is a way of easing the commercialization of a new technology at the crucial sticking point—the manufacture of new equipment—by using components already in production.

As another form of insurance, early versions and demonstration facilities are commonly designed with considerable reserves to cope with disappointments and to permit subsequent increases in capacity once operating experience is acquired. The first 500 KV line was operated originally at 400 KV; only later were experiments made to see if it could operate at 500 KV. As another example it is reported that the original design for the 1,000 MW RBMK nuclear reactor was so conservative that the upgrading to a 1,500 MW version can be achieved just by intensifying the heat exchange process, without modifying the design of the reactor as a whole. Likewise, in the VVER nuclear reactor, capacity is to be raised by exploiting the technical slack in the original model revealed through operating experience, rather than by extensive redesign of the reactor ( Atomnaia nauka i tekhnika v SSSR, pp. 37, 43). But even with this kind of insurance, Soviet experience with new power generating blocks (especially in boiler design), the MHD facility, the nuclear power program, and several of the excavator models, all suggest that, in relation to the actual capabilities of the R and D establishment, planners decided on big enough steps into the unknown to cause real difficulties.

I believe that the R and D approach outlined above has some serious disadvantages. U.S. observers of the MHD program seem to agree that the Soviet method is wasteful, since it fixes too many parameters in advance, puts ideas into metal and concrete too soon. And, as mentioned in Chapter 7 in connection with development of turbine compressors, Soviet commentators themselves are often highly critical of the failure to do enough testing as new equipment is being designed and readied for production. Certain aspects of the approach, however, may have important offsetting advantages, and the point here is not that Soviet R and D practices are inept, but that they are different from ours. It seems to me fairly obvious that the most helpful explanation for the differences from U. S. practices is the absence of the competitive pressures that influence R and D efforts in the market economy, especially in the private sector. In the United States, one of the main reasons for so much caution in the design and testing phase is the desire of the developer of new equipment that it be competitive and appealing to prospective customers when it is ready to be marketed. Potential users will want to know when it can be delivered; the seller of the equipment must be able to assure buyers about cost and performance. Failure here can mean that a firm’s innovation will not win acceptance. The same argument applies to a considerable extent even to the energy R and D programs sponsored or financed by governments rather than profit-making firms. In most market economies any new technologies developed with the stimulus of government financing will still have to meet market tests if they are to be successfully commercialized. Moreover, a technology brought successfully to the commercial stage in a given national context can be wiped out by international competition even if it is technically successful. As spectacular examples, consider the Concorde and the Dragon reactor.

In the Soviet setting many of those considerations do not matter. The user ministries have much more limited motivation and freedom to reject technologies that are not commercially appealing than do firms in a market economy. If the decision from the top is that the new technology is to be developed in variant A rather than variant B then there is enough monopoly in the system to foreclose the subsequent emergence of a truly superior form B to undercut the effort invested in A. Considerations of international competitiveness also appear much less forcefully in the USSR, and in a different way. Foreign competition is unlikely to supersede a domestically developed technology, except possibly in the form of technology transfer if the domestic program is a flagrant failure. International competitiveness may also influence the fate of the new technology if its development is motivated in part by the hope for foreign sales of the equipment (as in the example of power-generating equipment). This kind of international sensitivity was generally unimportant in the USSR in the past, though it has become more prevalent in recent years. There is now far more frequent resort to technology transfer; the USSR now has had some success and has larger ambitions regarding exports of energy technology. Eastern Europe, which in the past was essentially a captive market for the USSR, is now a much more discriminating comparison shopper in choosing the source of new energy technology. Students of the Soviet-type economies have often remarked on the potential virtues of opening up the Soviet economy as a way of disciplining domestic producers, and we have here an interesting variation on that idea—i.e., that more involvement with the world economy has the potential to transform somewhat even the R and D processes that precede production.

Another way to interpret this difference in Soviet and Western approaches to R and D is to relate it to differences in the kinds of uncertainties that dominate the respective systems. It was suggested earlier that many of the failures and delays noted in the various programs flow from the network of interdependencies, in which the various actors find themselves with very limited power to command information or to influence behavior of those on whom they depend in carrying out their part of the effort. In the United States, R and D program managers work on reducing technological uncertainties by precommercialization forms of R and D, the better to be prepared to deal with competitive uncertainties involved in winning acceptance at the commercial stage. In the USSR, the uncertainties imposed by interorganizational unpredictability are the ones that can doom a project, and program managers try to ease these by demanding only conventional inputs from outside the program. At the same time they are much less worried about getting the technology accepted once they have developed it, and so they push ahead with a crash program to deliver it in some kind of form, not worrying that they will have to do considerable backtracking afterward to cope with defects that emerge in practice and that were not thought about enough in the precipitate rush to commercialization.

The prospects for significant improvements in the interorganizational obstacles to improving Soviet R and D through more lateral interaction and responsiveness are probably not great, and the likelihood that the distinctive R and D patterns we have seen will retain their vitality is high. Many experiments have been tried to make R and D organizations more responsive to the needs of clients and to regroup organizations into new aggregations that would break down internal barriers. The shift to unified fund financing and to the creation of scientific-production associations (NPO) are examples of these efforts described in Chapter 2. But we should not be misled into believing that behavior will change significantly or that on balance the improvements from new structures will much exceed the defects that they bring with them. An official of the Soviet Committee for Science and Technology (GKNT) well acquainted with R and D in the electric power sector asserts that after five years under the new unified-fund system, though it has been incessantly lauded in the general literature as a gigantic step forward, there was no substantial change in the way R and D was performed and that the system should be abandoned (L. A. Vaag in Voprosy ekonomiki, 1975:8, pp. 121–122). In an effort to break down departmental barriers in the R and D process and increase sensitivity to client needs in the oil field equipment field, one of the scientific production associations called Soiuzneftemash has been created. Combining R and D organizations, experimental production and test facilities, and production plants, this organization also includes units whose responsibility is to follow up machinery deliveries with field service and testing. It sounds just like what the critics of Soviet R and D have recommended, but it is interesting to find the leaders of an R and D organization within Soiuzneftemash complaining that in fact the new arrangement has blocked effective work on new equipment:

The Association has begun to assign small urgent tasks to the Institute which have no direct relation to either the Institute’s planned program or specialization. They are mainly technical problems arising at different plants of the Association, which should be dealt with by plant technical personnel. Since January 1976, the Institute has received almost 600 such requests. It has been almost deprived of its experimental base, which is being largely used for mass production. Its research is seriouly hampered, and it is being turned into a maintenance brigade [Sotsialisticheskaia Industriia, 2 February 1977, p. 2]

One of the interesting questions is how to explain variations in R and D effectiveness within the energy sector, such as the difference between the electric power field (where the R and D process for all its faults has solved many significant problems) and the coal industry (where lavish R and D resources have apparently been powerless to raise the technological level anywhere near that in the capitalist countries). Several differences are probably important in the explanation. Relative resource scarcities may have played a part. In the coal industry much innovation involves capital-for-labor substitution, and an abundant labor supply simply meant little pressure to innovate in this way. In the electric power sector, a need to save capital and fuel may have spurred innovation. As suggested in Chapter 3, fuel is relatively expensive in the USSR, and since electric power generation is the major fuel consuming sector, it has naturally stimulated serious R and D effort aimed at cutting the heat rate. Conditions in the electric power industry also offer large scope for following a few simple strategic principles to raise productivity, an approach toward which Soviet planning is very partial, and which is easily seen in the programs we have described for this industry. It is more nearly possible in electric power than in coal to design a few basic pieces of equipment, such as a 300 MW power block to be installed in a standard building, and then to replicate this design all over the USSR. In the coal industry separate solutions are needed for thick and thin seams, for flat-lying and sloping seams, for underground and strip mining, and so on. But part of the explanation must lie in the kind of market-power considerations discussed earlier. Minenergo has a more powerful position in relation to its technology suppliers than do many other organizations in the energy sector. Most of its suppliers work for Minenergo almost exclusively, whereas many of the plants supplying equipment to producers in the other energy branches serve a bigger variety of customers. I also suspect that Minenergo has an unusual situation in its large project-making organizations—it has a very powerful say in the design of nuclear power stations, for example, and hence a strong influence on nuclear R and D organizations in machinery industries that supply the machinery that determines their technological level.

It might be thought that an in-depth look at Soviet experience in dealing with energy R and D might generate some predictions as to how Soviet energy policy will evolve, how technological considerations will influence outcomes on the big choices confronting Soviet energy policymakers mentioned throughout this book.* In fact, such a hope was a part of the inspiration for this study. As my investigation has proceeded, however, that expectation has become illusory, and I would like to conclude with some summary reflections on the general issue of how energy R and D interacts with energy policy.

It seems unlikely that the USSR will achieve the kind of dramatic breakthroughs in any given technological area that could tilt the balance of economic advantage decisively one way or the other on any of the major choices. Most Soviet energy R and D programs drag out over long periods, the equipment and facilities that finally emerge from them only partially meet the needs that motivated them. This is inherent in R and D, of course, and might be said about any economy, but I read the Soviet record as more disappointing in this respect than the norm in market economies. In the light of past experience, it just doesn’t seem realistic to expect the kind of great success in developing the technology of coal conversion or in power transmission that would give Kansk-Achinsk coal a dominant place in the solution of the energy deficit in the European USSR. Nor can one expect that the nuclear program, even in the light of its efforts to adapt reactors to permit applications other than power generation, will proceed, in accordance with the hopes of its managers, to make a dominating contribution to solving that problem.

By a similar argument, we should not expect that Soviet R and D in the energy sector is likely to so influence the structure of relative advantage in energy options as to move the main features of Soviet policy much away from that in the rest of the world. There are and will continue to be many distinctive features in the pattern of production and utilization of energy resources in the USSR, but these grow much more out of the nature of the Soviet resource endowment and demand structure than out of special Soviet strengths or predilections in the creation of energy technologies. It is true that Soviet planners have made some major commitments in energy technology that differ from those in the United States, such as the emphasis on the breeder reactor or the dominant place given the turbodrill. But it is still too early for a conclusion that they will succeed in making the breeder (especially when we remember the problems it entails in other parts of the fuel cycle) an environmentally viable and economic contributor to overall energy supply. And the turbodrill illustrates the phenomenon, fairly common in Soviet energy R and D experience, of pulling back from a distinctive technological line if it runs counter to world technological trends.

From the other side, none of the technological obstacles associated with the various energy policy alternatives are likely to be so insurmountable as to render a given option impossible. If domestic R and D efforts are unavailing in finding an acceptable solution to some crucial aspect of a given energy option (say in raising the productivity of gas pipelines), technology transfer often offers a way out. Our study suggests that that choice will be made only under the pressure of extreme urgency, but it offers a flexibility that heavily undercuts the notion that differential success among technological areas will be determinative in resolving the big issues of energy policy.

One of the distinctive elements differentiating the world view of economists from that of other students of social processes is a sensitivity to the fact that there are always more alternatives than appear to be available at first blush, that there are all kinds of substitutions and alternatives that make it very short-sighted to draw conclusions about what can and cannot be done. What is fascinating about R and D, what makes it so elusive to analyze from the point of view of making economic choices, is that it expands the range of choice still further. Technological alternatives aren’t given, they are created, and the way they are created is through R and D efforts. It does not make much sense to say that the breeder reactor won’t work, without specifying the particular technological embodiment of the general physical principle involved. Because R and D in its broadest concept involves the possibility of going back and starting with more fundamental approaches to almost any problem, the range of alternatives at any one place along the continuum of technological possibilities can be increased by shifting the effort backward toward the basic research end of the spectrum.

The following generalization about Soviet energy R and D may be helpful as we try to predict what may happen as Soviet energy policy unfolds. What is distinctive about the Soviet system, what constitutes the real weakness in the R and D process (which is one of the means to be manipulated in any large policy decision area) is its general clumsiness in its movement back and forth along the R and D spectrum—it makes commitments too soon, generates too few technological alternatives at each stage, finds it difficult to move backward to correct the consequences of miscalculations once commitments are made. This tendency is deeply ingrained in the system, perhaps an inherent feature of its basic organizational properties, and has a pervasive inference on technological advance. An awareness of that tendency is perhaps the best general orientation this study can provide for our continuing effort Lo evaluate technological potential as we seek to forecast Soviet energy policy.

____________

*For a more extensive description of the choices inherent in the Soviet energy situation, see the excellent treatment in Shabad and Dienes (1979).