The Future Market for Finished Intelligence
APPROVED FOR RELEASE 1994
CIA HISTORICAL REVIEW PROGRAM
2 JULY 96
The production of finished intelligence is a principal purpose of all U.S. intelligence activities; neglect of it is unacceptable for the future. Senate Select Committee, Final Report ... with Respect to Intelligence Activities, 1976.
THE FUTURE MARKET FOR FINISHED INTELLIGENCE
That finished intelligence will be neglected in the future by either producers or consumers seems highly unlikely, but that greater professional challenges lie ahead for us in producing a satisfactory quality and array of finished intelligence seems inevitable. In meeting this challenge we will need, among other things, to think carefully about the market for which we produce.
This study, undertaken by the Center for the Study of Intelligence at the request of the Deputy Director for Intelligence, reviews the present market for CIA's finished intelligence product, seeks to forecast2 the direction of change in that market over the next five to ten years, and suggests steps that might be taken to prepare for these changes. Our assumption is that the nature of CIA's finished intelligence product is and will be largely determined by the shape of the intelligence market. We see this market as twofold: (1) as to the number and variety of organizations and individuals who request, receive, or make use of CIA's finished intelligence product; and (2) as the nature and scope of their major substantive interests.
In the main, the study concludes that:
— tomorrow's market will continue to be dominated by the traditional national security consumers — the Chief Executive and his senior policy makers — and by traditional issues such as Soviet capabilities and intent;
— but reaction times for U.S. policy makers will be shorter, and the pressures for more and sharper advance intelligence warning will grow;
— similarly, crisis monitoring will take on greater effort and scope. New types of crises which are basically economic or environmental may be as significant for U.S. security interests as the military and political crises of the past;
— what we produce and how we produce it could be strongly influenced by organizational changes in the intelligence community itself. But in the market place, CIA will continue to be looked to for finished intelligence in all major disciplines and on all major subjects;
— thus, the pressure to do more, but probably with much the same resources, will continue, and will force us farther in the direction of organizational flexibility, greater analytical efficiency, and more stringent criteria for choosing what we do or do not produce.
II. Evolution of the Intelligence Market
A number of factors have significantly influenced the evolution of the market for CIA finished intelligence from past to present. Of these, the most important appear to have been:
— changes in the substantive focus of consumer requests for finished intelligence, stemming principally from policy maker perceptions of the main threats to U.S. national security;
— improvements in the intelligence collection and analysis process which have shaped the market by enabling us to examine old issues in new ways, or to develop new products on new subjects — which in turn has resulted in additional consumers;
— organizational changes in the intelligence community which led to CIA being formally charged with additional substantive responsibilities, with the result that CIA took over an existing market or moved on its own to build an analytic and production capability where the existing product seemed unsatisfactory to consumers;
— deliberate resource shifts within CIA whereby the Agency dropped out of certain markets as a result of modifying or closing out existing products, or massed resources in such a way as to increase the diversity of intelligence products and thus broaden the market;
— a gradually accelerating market interest in immediate or quick analysis-current intelligence — which has had a great impact on the overall posture and focus of CIA analytical resources;
— concurrent with the growth and interest in current analysis, a growth in the market for in-depth and integrated analytical work, partly the result of the increasingly perceived complexity and interrelationship of international affairs, and partly the result of the improvement and sophistication of analytical techniques and intelligence capabilities.
How these market factors influenced production is seen in an historical review of the evolution of CIA finished intelligence on military, economic, and political subjects.
When the Agency started off, Communist expansionism was viewed as the main threat to U.S. national security. This was the Cold War, and the easy identification of our adversaries and of the threats posed by them provided a long period of relative stability in the key subject matters of international affairs and a similar stability in the collection, analysis, and production of foreign intelligence. In this period, the foreign intelligence effort was aimed almost entirely at the Communist countries. Within the intelligence community, CIA was responsible mainly for the production of finished economic intelligence; production of intelligence on political and military developments was primarily the responsibility of the Departments of State and Defense.
By the late 1950s, the market had begun to change. The launch of Sputnik 1, Soviet development of intercontinental weapon systems, and the subsequent bomber and missile "gaps" created new consumer demands for intelligence on Soviet strategic capabilities. The result was a considerable enlargement of CIA's military intelligence effort, especially in the area of Soviet strategic forces. In 1961, this trend was accelerated by the inauguration of a President who had charged during his campaign that the U.S. was failing to meet the challenge presented by the growth in Soviet strategic capabilities. His Secretary of Defense looked in part to CIA for intelligence judgments on Communist military developments; the new IDCI was determined to satisfy him and to provide the President with CIA's independent views on Soviet and Chinese military affairs. Largely in response to this combination of increased substantive priority and new consumer demand for military intelligence without presumed Department of Defense bias, a Military Research Area was established in the DDI's Office of Research and Reports (ORR), and a Military Division was formed in the Office of Current Intelligence (OCI).
At about the same time, the advent of satellite photography, and the new analytical opportunities which this technological innovation provided, greatly enhanced the intelligence community's ability to assess foreign military developments. The increase in the number and variety of issues which could now be authoritatively addressed led to further changes in the intelligence market by increasing the number and variety of customers for CIA's military and technical intelligence products. By 1963, a new Directorate had been formed within CIA, responsible both for developing new technical means of collecting intelligence information and for analyzing and producing intelligence on foreign scientific and technological activities. Still later, the disparate military intelligence components of the DDI were combined into a new Office of Strategic Research (OSR).
By the late 1960s, the combination of improved technical intelligence collection methods and the evolution of a situation of rough equivalence in the strategic strengths of the U.S. and the USSR significantly increased the prospects for arms control. This in turn created new market demands for military intelligence and new pressures on CIA and the rest of the community for in-depth analysis and reporting on Soviet and Warsaw Pact military forces. A whole new infrastructure of analytical groups and intelligence publications evolved over time to support the various arms control efforts, such as the Strategic Arms Limitation Talks (SALT), Mutual and Balanced Force Reduction (MBFR), and the test ban treaties.
Still more recently, in response to the increased demand for more in-depth knowledge of Soviet and Chinese military doctrine and strategic policy, a Strategic Evaluation Center was set up within OSR. The new sources and methods had enabled us to progress over time from the collection and analysis of basic, factual data on Communist military forces — i.e., the accumulation of bits and pieces of intelligence information — to their manipulation and integration into considerably more sophisticated, in-depth analyses of Communist military doctrine and capabilities.
We can thus regard the past evolution of our effort in the production of military intelligence as a product of changing substantive interests on the part of consumers, improvements in collection and analytic techniques, and changing organizational arrangements. The trend, it will be noted, was consistently toward more depth and complexity in the product.
Similar though less extensive changes took place over time in CIA's economic intelligence effort. The market for economic intelligence, always a primary concern and responsibility for CIA, has been influenced mainly by changes in the world scene.
Resource limitations elsewhere in the community also had a significant impact. At the outset. CIA concentrated its economic intelligence resources on the Communist states. In 1961, however, the Department of State's Bureau of Intelligence and Research (INR), which had been following economic developments in the non-Communist world, dropped most of this responsibility, leaving it to CIA. The effort on nonCommunist countries intensified (and was institutionalized by formal agreement with State) later in the 1960s with the development of African independence movements, Soviet economic penetration of the less developed countries, and the economic growth of Japan and Western Europe. In the 1970s, the national security implications of the decline of the dollar, the subsequent balance of payments crisis, and the still later oil and energy crisis brought new consumers, new demands, and new production opportunities to CIA's Office of Economic Research. OER responded with new publications and numerous changes in its organizational structure. Here the trend was toward more current analysis of issues with a broader substantive and geographic range.
In political intelligence, the changes have been less dramatic, though affected by a similar interplay of internal and external influences: consumer preference, changes in the world scene, and resource considerations. Originally, political intelligence, like military intelligence, was a secondary concern of CIA; primary responsibility rested with the Department of State. In 1961, however, resource limitations caused the State Department to drop responsibility for in-depth analysis of political developments in the non-Communist countries. A modest increase in CIA's political research effort was undertaken the following year with the establishment of a Special Research Staff in the DD1, but most of CIA's political research was still concentrated on the Communist countries; analysis of political developments in the non-Communist world was largely limited to OCI's current intelligence reporting and to the biographic research effort in the Central Reference Service (CRS).3 New, more sophisticated methods of analyzing political developments were introduced in succeeding years, but it was not until 1974 that the need for in-depth political analysis on a broader geographic basis was perceived to be great enough to warrant establishment of an independent Office of Political Research (OPR) in the DDI.
We conclude from examining the past evolution of the organization of CIA for the production of finished intelligence that it not only reflected outside market factors — such as shifts in substantive interests and changes in community organization — over which the Agency had little control, but that the ways CIA itself went about its business — new analytical techniques and collection methods — influenced the market as well. In turn, this market evolution required new CIA resources, organization, and products. One might characterize the overall trend in past market factors as one which led to increasingly diverse and complex products which called for increasingly sophisticated analysis. No slackening of this recent market trend is evident. In fact, it seems to be accelerating and, as we shall see later, appears likely to continue into the future.
Today, the intelligence market continues to be dominated by the original or traditional national security consumers — the President and member agencies of the National Security Council. Intelligence as a whole continues to be dominated by the effort to collect, analyze, and produce intelligence on the strategic capabilities of the USSR and other denied areas (consuming, as it does, about 75 percent of the community's resources annually). But, as we have noted, it is a far wider-ranging and more varied product on the denied areas today than in earlier years. The remaining intelligence effort, moreover, is also far more varied than in earlier years. New substantive intelligence issues have created new markets for CIA's product without reducing the importance of old issues or of old markets.4 It is this expansion of the market that has created the key management issue for the future: that of doing more with less. This will be the prime problem for the producers of finished intelligence in CIA under any likely scenario of future influences on the market.
III. Some Aspects of the Future
Many of the factors which have already begun to influence today's market are likely to be with us five or ten years hence. Some of these, however, are more likely to grow in significance than others, and it is to these factors that we now turn.
Future Substantive Issues
In the substantive area, predicting the shape of the world to come is well recognized as a hazardous business. Basically unpredictable events can have a major influence on the course of international relations and on the attention which we in intelligence must give to them. Who would have predicted 10 years ago the priority of international economic and energy issues, terrorism, and illicit drug traffic as key intelligence questions? Yet some elements of continuity as well as change can be identified with considerable confidence. The following discussion concentrates on those issues which we believe are likely to have the most significant impact on the way we in CIA must operate to serve our main consumers in the coming years.5
No diminution of consumer interest appears likely in the next few years on the traditional problem of assessing the military capabilities and intentions of Communist countries, even if the U.S. and the USSR remain committed to detente. It seems probable that the situation of rough equivalence in Soviet and U.S. strategic forces will be maintained — with or without new arms control measures — and that this will require extensive intelligence support. There is a good likelihood, moreover, that events will transpire to increase above even present levels the importance of timely and accurate intelligence on Soviet and Chinese activities.
If U.S. military disengagements abroad continue at the scale of the past few years, as the economic as well as political omens seem to portend, the reaction time needed to prepare any military response to future Soviet and possibly Chinese initiatives will increase and will necessitate even more advance intelligence warning in order to develop appropriate and timely counterefforts. We will also have to be watching the initiatives of potential Soviet and Chinese agents of involvement, such as Cuba and Vietnam. New analytic and collection resources (e.g., day-night imagery and new analytical modeling techniques) will aid in increasing the thoroughness and sophistication of our finished intelligence on these subjects, but as we have discovered with past technical and methodological improvements, they are also likely to lead to an expansion in the number of our consumers. and in the nature and substance of their requests. There thus would appear to be little chance for reduction in the production resources devoted to traditionally important intelligence subjects over the next few years.
With no lessening of interest in our traditional areas of intelligence concern, events in both the industrialized and less developed nations in the non-Communist world will take on new significance as intelligence problems. These problems will be increasingly transnational and interdisciplinary in nature — a conclusion supported by almost all those consulted in connection with this study. What once would have been perceived as primarily a European or Asian economic problem may become predominantly political in nature or take on new and added texture involving other disciplines. Consumers interviewed for this study were reluctant to be specific about areas or subjects of key intelligence interest for the future — beyond the traditional ones — which leads us to conclude that CIA will get no more definitive help in the future than in the past from its consumers in defining or circumscribing the proper substantive bounds of the market for finished intelligence.
Those queried in connection with this study agreed that today's trends in international relations indicate that the intelligence and analytical environment of the late 1970s and the 1980s will be much more complex. From their remarks and our own study, we believe that this environment will be mainly characterized by:
— continuing fragmentation and increasing diversity within the Communist and non-Communist worlds;
— increased interdependence between the U.S. and the non-Communist industrial countries on military, political, and especially economic matters;
— increased interdependence between the developed and the developing nations;
— increasing significance of economic resources in international political and military affairs;
— increasing importance of technology as an element of economic and political as well as military power.
Crisis monitoring, another key element of the finished intelligence process, will probably take more effort, and thus more resources than at present. Predicting crisis locales and issues, and identifying which ones might be amenable to U.S. action, will be an even more intractable problem. In this connection we note:
— the broadening of U.S. national security interests beyond the traditional political and military concerns to such areas as energy, the environment, and food. (A future drought in Western Europe, for example, could impose on us the same crisis reporting load as a political coup in Portugal.)
— the strong possibility of an uncontrolled spread of nuclear weapons, thus greatly heightening the stakes and risks of a crisis, and the need for accurate and even more premonitory intelligence on likely crises.
— the rise of international terrorism to new levels of sophistication and danger, and the difficulty of predicting where it will strike next.
The extension of crisis monitoring beyond the traditional political and military reporting realms to the disciplines of the economist and scientist will call for a more integrated intelligence approach, and this can only heighten the demands for managerial as well as analytical resources devoted to the effort. Experience has shown in recent years that crisis reporting is consistently one of the best received and most highly complimented of the Agency's products. Thus, in line with the general evolutionary development of the intelligence market of the past — in the direction of product utility — no slackening of consumer interest in crisis reporting can be expected. (Note the recent pressure for a community-wide situation report system in crisis monitoring.)
Not only will the international environment be more interdependent and complex in the coming years, thus bringing changes in the market for intelligence, but there is a strong possibility that the market may also be additionally shaped by major organizational changes in the executive agencies of government and in the intelligence community.
Changes in the Executive
Although the substance of international relations has undergone considerable change since the early 1950s, the basic structure of government for conducting U.S. foreign affairs has remained largely the same. With international commercial and financial relations becoming more central to foreign policy considerations, for example, the Treasury Department is still absent from the National Security Council. (It was, however, brought into the intelligence community and given representation on the USIB.) Bureaucratic jealousies and the difficulty of obtaining statutory authority for significant change in the national security structure have worked against such change. The result has been a proliferation of ad hoc committees, councils, boards, and agencies to handle the increasing number of interdisciplinary problems which otherwise would fall between the jurisdictional cracks of the existing structure, and most of these new entities have become consumers of CIA's product.
For the future, the national security structure appears likely to develop along one of two lines: continuation of the present ad hoc approach, or else extensive change. The latter is most likely to come about with an administration in which the Executive and Legislative branches are led by the same party and the inauguration of a President dedicated on principle to changing the present organization of government.
The extent to which any President could successfully alter the existing national security organization is, of course, open to conjecture. It would vary with the amount of support he could obtain from Congress and from the bureaucracy itself, but two trends are likely to dominate any such change in the near future: expansion of the national security apparatus to include additional governmental departments; consolidation of the present structure of ad hoc and sometimes overlapping components dealing with various economic and resource issues (e.g., the Energy Research Development Administration (ERDA) and Federal Energy Administration (FEA); or Council on International Economic Policy (CIEP) and Economic Policy Board (EPB). One result of the combination of these two factors or trends could be the formation of a new Department of Foreign Trade and incorporation of the new department into the National Security Council. For CIA, this would mean absorption of some of our present customers into a new cabinet-level consumer of foreign economic intelligence, and possibly the creation of a new departmental producer of such intelligence. (The possibilities for the creation of new departmental-level intelligence units are examined in more detail in the next section.) The end result presumably would be a more manageable and coherent market for economic intelligence and, over the longer term, the absorption of at least some of CIA's present analytical and reporting responsibilities by the new department.
Without reorganization of the present national security structure (and perhaps even with such a move) there probably will be a proliferation of Executive Branch components dealing with the national security aspects of international economic,environmental, energy, civil technological, and resource issues, and each new one would become part of the intelligence market of the late 1970s and 1980s. Such a proliferation would inevitably add to the managerial complexity and the drain on managerial resources in producing finished CIA intelligence, since it would at least mean an expansion of requests from new customers, even if the general subject matter remained the same.
Whatever the election outcome in November 1976, there will be at least one new President during the next ten years because of constitutional limitations. In the past, such change has always imposed its own particular geometry on the market for CIA's product, often for a period of months or even years. In addition to making some changes in the executive structure, each new administration brings many new officeholders to the national security arena with varying levels of knowledge, differing preconceptions, and their own ways of doing things. This, in turn, creates a requirement for CIA and the rest of the community to adapt to the style and desires of the new personalities and often to offer basic education on those substantive issues on which intelligence can contribute. This essentially two-way familiarization process usually has a substantial impact on the use of CIA resources and managerial time. Also, new administrations usually consider that they have a mandate for and the freedom to undertake significant foreign policy initiatives at the outset of their term of office. This usually requires us to produce more in-depth educational material on issues of interest to the administration, as well as more estimative products.
Whatever the changes in the executive branch, the President and the NSC will continue to be the primary consumers of national intelligence, which CIA is primarily responsible for producing. The NSC principals and their staffs will, therefore, continue to be our most important customers.
Change Within the Community
The possibility of the creation of additional departmental analytic intelligence units in the economic area, or in other government spheres such as agriculture, energy, or commerce, is worth further thought. Theoretically, the creation of such units within new or existing departments would be a substantial factor in both the shape and the supply of the future intelligence market. Upon reflection, we do not rate as very high the possibility of the creation of such new units, at least over the next few years, and even if it comes about, we think it would mean only a rather limited reduction in the CIA analytic load. Our reasoning is a follows:
— other government departments currently have the same strains on resources that are felt in CIA and wish to channel any surplus into their mainline responsibilities. This is not likely to change.
— intelligence as a focus of departmental activity is not seen conceptually by Treasury, Agriculture, Commerce, or ERDA as central to their operations.
— the present philosophical tendency in government is to support the importance of "competing centers of analysis." This is reflected in recent congressional criticism, while the recent executive directive on intelligence does not even mention substantive intelligence coordination as such.
Even with the creation of some new departmental intelligence unit, we think the pressures still would be heavy on CIA for production of national-level intelligence on, for example, economics or energy issues, just as they have remained heavy on us for political and military issues with the existence of INR and the Defense Intelligence Agency (DIA). Most of the consumers interviewed impressed upon us the degree to which consumers will look to CIA in the future to provide an independent interpretation of national interest issues. The more one talks to consumers, the more one is impressed by their belief in the need for an independent CIA voice in the major analytic fields. Even if a vigorous new departmental intelligence effort were undertaken, the start-up process would run several years at the least before a really competitive, full-spectrum product could be developed in any major area. It would be 1980 or so at the earliest before a full shift in our responsibilities could take place.
While the chances for any wholly new and competitive departmental intelligence unit may be small for the next few years, there do appear to be sound grounds for believing that pressures will move in the direction of stronger intelligence community management of efforts to produce a better overall product through more rational use of existing community resources. This will inevitably have a bearing on the shape of the market for CIA's product. Presently, this particular community activity is centered in two areas: the product evaluation process of the Intelligence Community (IC) Staff, and the National Intelligence Officer (NIO) system. So far, the product evaluation process has not had a major impact on the shape of the product, but the establishment of the precedent provides a significant potential channel for registering and regulating consumer requirements on CIA. Over the long term, given support by the Executive and by management, it could become a much more potent tool for regulating the shape of finished production by the Agency and for rationalizing the overall community production effort.
The NIO system is already playing a role in this effort. It serves a severalfold need: improved communications with the diverse consumers; improved coordination of the diverse community product; improved use of the diverse community analytic assets. Indeed, the Deputy to the DCI for National Intelligence Officers is specifically charged with ensuring that consumer requests do not overload the system, that realistic production priorities are established in consultation with consumers and producers of the NIO product, and that nonproductive overlap in reporting responsibilities within the community is eliminated. The direction of community policies as related to finished intelligence production is many-faceted, so the full impact on CIA and the market can be only dimly perceived at present. We think that the major community policies will go in the direction of:
— obtaining more complete control over total community intelligence expenditures; this will result in further pressures on the Agency to concentrate analytical expenditures on areas of demonstrably important consumer interest.
— further community-wide, joint identification of intelligence production priorities going beyond the present Key Intelligence Questions (KIQs) and other guidance techniques toward explicit community directives as to who will concentrate on what.
— a community-wide ordering of intelligence publications so as to avoid overlap and duplication.
— more community-directed evaluation of the quality of the CIA finished intelligence product.
— more inter-Agency analytic teaming of a multidisciplinary nature on important regular as well as ad hoc products.
— a more concentrated community effort in the production of finished intelligence during crisis situations.
What impact will this have on the market for intelligence and correspondingly on CIA's response to that market? It is to be hoped that it will eliminate some of the product duplication criticized by consumers in the past. By eliminating duplication, some improvement in the quality of the product may result. What needs to be kept in mind, however, is that both by executive directive and by quality of product, the CIA is being looked to increasingly as the main producer in all areas of national-level intelligence. It is unlikely that this will change in the foreseeable future. Therefore, budgetary and other restraints that may be imposed on finished intelligence production on a community basis in the next few years are not likely to cut off many of CIA's present production responsibilities. They are most likely to fall, if at all, on present finished production at the departmental level which duplicates national-level treatment. CIA will have to do as much as it does now; it just might not have as much departmental competition.
Do we mean by this that the coordination of the substantive intelligence product between ourselves and the other producers will dwindle? No, probably not much below the present overall level, which is reduced somewhat from high points in the past. It is unlikely that a new drive will emerge in the coming years in the government for the production of "consensus" intelligence, inasmuch as the record now is well documented and publicized that consensus often contributes to intelligence failures. Thus, coordination of substantive products is more likely to take the form of interagency discussion and cooperation in production, with no requirement for total agreement or consensus. How will this affect the intelligence market? It will simply add to the climate of expectation by consumers that CIA should produce across the total spectrum of intelligence.
The Legislative and Nongovernmental Markets
Another new influence is the tendency in some quarters to view intelligence as a product with a legitimate market beyond the confines of the executive agencies, possibly extending even to the public. The reasons for this are complex and beyond the scope of this paper, but the trend is likely to continue in the future.
An example can be found in the congressional market. The new executive order officially certified this market, naming the DCI to "act as the principal spokesman to the Congress for the intelligence community and facilitate the use of foreign intelligence products by Congress." The number and variety of requests for substantive intelligence information from Congress have been expanding, partly as the result of the creation of new issues, partly as the result of increased congressional efforts to wield more influence in the foreign policy field, and partly as the result of the "discovery" of CIA by various members and committees of Congress during the recent investigations.6
Over the next several years, the market for intelligence in Congress is likely to be influenced by several conflicting factors. Some that influenced the recent expansion in this market will still be at work — e.g., the "discovery" of CIA by various congressmen and their staffers, and continued growth in congressional concern for a wider variety of foreign policy issues. On the other hand, CIA will have a lower profile on the Hill as the large-scale congressional investigations of intelligence fade into history. There will also be additional instances in which executive-legislative differences over foreign policy issues will result in executive limitations on the CIA provision of intelligence information to Congress. Moreover, if Congress and the Executive Branch come to be led by the same party, Congress may be less inclined to do battle with the Executive and, therefore, be a less avid consumer of intelligence. But the next few years are also likely to see more interest in consensus foreign policy; and the Executive may want CIA to work closely with the legislature, especially in those instances where the available intelligence might strengthen the case for a new foreign policy initiative by the administration.
In any case, security considerations will continue to inhibit the full disclosure of intelligence information to the Legislative Branch. No fully satisfactory means have been found for supplying ever larger amounts of sensitive written intelligence to Congress. What is likely is a growth in verbal intelligence inputs in the form of briefings of congressmen and committees. Such support fits closely with the way Congress traditionally absorbs information and mitigates the effect of the committee and staff system which impedes the access and handling of written intelligence information in Congress. On balance, we would predict only a limited growth in the next few years in the congressional market. While the demands on senior officials of the Agency for briefings will continue to be heavy, this process is not likely to have drastic impact on the total resources of the CIA, since most of the intelligence supplied will be redo's of products already prepared for other purposes.7
To the extent that congressional demands for written products grow, we think that growth is likely to come primarily in requests for deep-research studies requiring considerable time, and probably an analytic team. Such tasks have already been requested of us on a number of issues and have had considerable resource impact in some parts of the Agency as Congress seeks to make itself more expert on various economic and defense issues. Requests like this could spread to other substantive areas and components.
The trend toward provision of intelligence to markets outside the Executive is also showing up in the availability of more unclassified finished intelligence for the public. There has been expanding interest in products which do not necessarily require the use of classified information. CIA atlases, for example, are available to the general public in tens of thousands of copies through the Government Printing Office, and various CIA reports on such things as basic economic statistics, the world population problem, food shortages, and climatic trends are available to the academic community and other interested parties.
This trend toward provision of intelligence to the public is likely to increase. Whether the task will result in any substantially greater load on our resources in the next few years is less clear. By and large, the service to this market has so far involved the sanitization of products already produced for other intelligence customers — a somewhat time-consuming but not overly significant load on production resources. The Agency is presently in a position to decide what and what not to make available on a regular basis. Some increase in the provision of finished intelligence production to the public could probably be undertaken for public relations purposes without extensive commitment of additional resources.
A presidential policy of using all the available tools of the Executive more openly in an effort to build a consensus on foreign policy could, of course, make a big difference. In that case, the public market might become a significant additional load on our resources. From the standpoint of the intelligence professional — for the sake of security and objectivity — the less intelligence products are used in this manner, the better. But a President facing demands for more openness in government, and determined to build support for his programs, may not view it in this light. By taking advantage of the trend toward decompartmentation and downgrading, and toward the increased availability of unclassified, declassified, and sanitized information, a determined Executive could place the Agency under strong pressure in future years to build a product deliberately tailored for public consumption.
One can discern in the tendency toward decompartmentation, downgrading, and declassification, a trend toward the more widespread use of intelligence generally — at lower levels within the existing market as well as in entirely new markets. The lowerlevel consumer, though often playing an important role in the early formulation of policy, will usually be more "departmental" than "national" in nature, and, thus, will often be able to be served by existing producers of departmental intelligence or by the spin-off from CIA's national product. But we might also begin to see such things as foreign economic intelligence being sanitized and made regularly available to U.S. business and commercial interests, either directly or through other government agencies. Freedom of Information Act requests probably will serve to accelerate this downgrading process. The significant provision to the public of finished intelligence in this way would seem highly conjectural at this point, however, and if it does develop, probably would not tie up a significant amount of production resources for some years to come.
An allied trend of the last few years has been the development of a market for finished intelligence furnished to foreign governments as a means of influencing the policies of those governments or of generating a more favorable environment for military or intelligence operations. Satellite photography is provided to the Commonwealth countries as part of a regular and extensive intelligence exchange program, and is made available selectively to West Germany and South Korea. It has also been used from time to time in briefing officials of at least two dozen other countries. Some expansion of this practice, perhaps involving development of individual finished intelligence products specifically tailored for the purpose, can probably be anticipated, but does not seem likely to become a major drain on resources, at least over the next few years.
The Tactical Intelligence Market
In the coming years, the already hazy distinction between national and tactical intelligence is likely to be further blurred by the aforementioned trend toward decompartmentation and the development of more rapid communications. One result of this will be that field commanders will become more frequent consumers of CIA's national product; conversely, the National Command Authority will make greater use of tactical intelligence, especially during crisis situations when tactical intelligence tends to take on even greater strategic or national significance.
The impact which this will have on the market for CIA's product is difficult to measure in the abstract. It is in any case likely to cause some expansion in the substantive scope of issues which we will have to address — into areas which might previously have been considered tactical in nature — and probably will create a demand for more analysis and comment on evolving situations as viewed from the perspective of people in the field.
New Analytical Techniques
As we have seen earlier, another influence which has begun to affect the intelligence market is the role of new analytical techniques in the production of finished intelligence. These techniques generally bring about a more structured, comprehensive, and less intuitive approach to analysis. Among the more recent examples are the Bayesian analysis experiments,8 the procedures developed for estimating agricultural harvests, and the system dynamics and other types of models developed for such tasks as assessing oil production in China. We think these techniques by and large have some common effects which are likely in the coming years to have considerable influence in the intelligence market and in the consequent use of analytic resources. In this connection, we note that:
— the new analytical approaches usually necessitate increased use of teams of analysts from across disciplines and organizations, thus complicating and increasing the managerial load in producing such intelligence beyond levels associated with more traditional methods.
— more individual analyst time is required in the new analytical approaches, with additional needs for meetings, coordination, etc., both inside and outside the Agency.
— inputs of data, opinions of experts, activities of contractors, etc., from outside the normal intelligence community are often required. This not only increases the time load on the analysts and on management, but widens the potential market for new methodological products as well.
— the new methodologies, models, etc., usually provide a technique and in some cases a requirement for continuing replication of the analysis (i.e., Bayesian analysis on the Mideast war possibilities) which can become an additional, routinized market product.
For the next several years, we see only a gradual growth in the use of more structured analytical methods. These processes do indeed seem to be the wave of the future in most analytic disciplines, however, and by their nature will pose substantial strains on our production apparatus, unless that apparatus is modified to handle the particular organizational problems they pose.
IV. Some Ways To Cope With the Future
Ironically, the problems of the future will be created largely by our success in responding to the demands of today— the better we do our job now, the greater will be the demand for our product and, thus, the wider our market tomorrow.
Despite the almost certain emergence of a busier and more complex intelligence environment in future years, personnel and dollar resources available to CIA and to the rest of the community are not likely to increase significantly; indeed, real dollar resources are likely to decrease in many components of the community. As noted in a preceding section, the paramount problem for management in the coming years will be how best to balance the demands of the busier intelligence market with the capacity to produce thorough, timely, and objective intelligence with limited analytical resources. With respect to resources, the balancing act will call for management to be more critical in determining such things as which issues are the most pressing, which are legitimate ones for CIA to be addressing, and which can be addressed effectively by other governmental or nongovernmental components. With respect to the market, management will have to be more circumspect about such things as which consumer requests to honor, and how much support should be given to Congress, to Executive Branch components outside the formal national security structure, to the business community, and even to the general public.
These sorts of questions obviously will be more difficult to answer in the new areas of intelligence endeavor — those outside our traditional concerns — than in the old areas where CIA responsibilities are fairly clearly delineated by charter and tradition. With economic and civil technological problems in the non-Communist world taking on increasing importance, it is natural to suggest that CIA expand its effort on these two topics, especially when the Agency has such a long lead in experience over other analytical components, at least on international economic issues. But there are significant risks involved in attempting to take on every new issue which comes along in the economic and civil technological areas. Not only is CIA limited by statute in the extent to which it can legitimately examine the leading role which U.S. firms often play in foreign economic and technological developments, but there is danger that by reaching farther into new areas of endeavor, CIA will find itself falling behind in fulfilling its primary responsibilities — i.e., strategic threat assessment and crisis monitoring.
This is not to say that CIA should shy away from new, less traditional issues; on the contrary, we will be looked upon to make important contributions in these areas. What it does say is that, without added resources, CIA must give much greater attention to husbanding its existing assets, and to finding more efficient ways of doing things. Flexibility has contributed significantly to CIA's success in the past. Its relatively small size, fewer bureaucratic restrictions, "can-do" attitude, and organizational slack9 have given CIA the ability to redeploy its resources rapidly in response to new consumer demands and new issues, without significantly undermining its basic analytical effort. In the future, we will have to preserve as much flexibility as possible in our existing resources, but as our resources are spread thinner, flexibility must increasingly be combined with selectivity. We will have to narrow the focus of our effort in some areas, drop lower priority projects, and stretch out some programs.
We will also have to get more mileage out of the work we do complete. Even though much of the Agency's most effective and influential work is done in the form of customized service, we probably will have to produce more multi-tiered and multiclassification reports so that the same product can reach a wider, multi-level audience. Finally, we will have to resist the temptation to satisfy a growing number of peripheral consumers (i.e., those outside the traditional national security arena) unless they can be served as a by-product of, or spin-off from, other production.
This is all well and good, it will be said, but how does CIA management translate these principles into specific criteria suitable for making specific decisions on resource allocation for the production of finished intelligence — particularly when the decision as to whether to produce or not is sometimes simply not within our control? On the latter point we have no answer, except to note: first, that the DCI now has explicit authority from the recent executive directive to ensure the "propriety of requests and the responses thereto" from the executive departments of the government to the intelligence community; and second, that the bulk of finished intelligence produced by CIA in the past has been self-initiated (although this has not been true for the most important individual memos, reports, and estimates). The percentage of self-initiated material cannot be pinned down with exactitude, but several sampling counts suggest that it runs to around 80 percent of our current and in-depth product in all analytical fields.
There would thus appear to be significant room for discretion on the part of management in tailoring resources to meet real consumer or market requirements. To be sure, a number of serious efforts have been undertaken to ascertain the specifics of these requirements, usually resulting in no enduringly useful conclusions. When asked, consumers generally tend to say they want it all. In view of the percentage of selfinitiated CIA finished intelligence, however, the best strategy for future years may well be to self-initiate the cuts and wait for the reaction.
As for the problem of developing criteria for management to measure the relative values of producing finished intelligence on new subjects and areas that will emerge in the years to come, the following guidelines may merit consideration:
— how much of the product of the undertaking will really depend on a unique intelligence informational and analytic input; how valuable as a contribution to the whole governmental understanding on the general subject will the specific report, memo, etc., be if it consists only of a unique intelligence input?
— in a related vein, have changes in classification of material (i.e., decompartmentation, more overt use of covertly collected information, and the availability of relevant unclassified information) eliminated the past rationale for analysis and production of the particular undertaking by CIA?
— is the subject currently treated thoroughly and well in professional journals and other media outside the intelligence community, and perhaps outside the government? (See, for example, Nathan Keyfitz's article on food and population trends in Scientific American for July 1976; this essentially multidisciplinary piece is generally comparable to the items which CIA has produced on these subjects so far.) If academia is doing a good job, is CIA input really needed?
— how directly does the suggested undertaking relate to concerns that are unequivocally within the bounds of admittedly expanded national security interests?
— if a consumer-requested undertaking, how important a role in the decisionmaking process on this particular national security issue does the requester play? (This may sound contrary to our deeply-held principles of service to the Executive, but it simply must be given more conscious consideration by management in the future if we are to do our best and make it count.)
The use of such criteria in deciding whether or not to undertake a particular project or line of finished intelligence production would offer CIA management a meaningful and responsible plank on which to stand in ordering its future use of analytic resources. An effort to develop more thorough and specific criteria of this type might well be worth the time.
The Problem of In-Depth Research
The broadened scope of CIA production in recent years and the increased emphasis on current intelligence, without a corresponding increase in analytical resources or decrease in effort on traditional national security issues, has had important side effects. Analysts and staff who formerly spent a significant amount of their time enhancing their own and their unit's in-depth understanding of the subjects within their ken, principally by conducting in-depth "basic" research (i.e., work not designed to support a specific finished intelligence product as such), find themselves continually deterred from such research by the need to produce more quick-reaction products. Inevitably, as resources are spread thinner in this way, in-depth research suffers.10
If the forecast in this study is correct, CIA's research effort is likely to be threatened further in future years by additional current reporting pressures created by a general acceleration of important international events, an increasing priority for a growing number of issues, and the increased availability of intelligence information in real-time —as events happen. The combination of these trends could seriously degrade CIA's research capability unless additional steps are taken to protect or isolate research programs from current intelligence demands. At present, current intelligence is handled differently for each of the three major disciplines within the DDI: political intelligence is handled in two different offices, military intelligence by different divisions within the same office, and economic intelligence by the same analysts in the same office. (Scientific and technical intelligence is handled yet another way in the DDS&T, but comes closest to the way economic intelligence is handled in the DDI.)
There are good reasons for handling current intelligence differently in each case, but the thinning out of resources and the increase in current intelligence pressures in the coming years may dictate other approaches. The approach that would go farthest toward protecting the research from the current effort would seem to be the one presently employed for political intelligence — i.e., one office or separate staff for carrying out in-depth research, and another for doing quick-reaction pieces and current reporting.11 This approach has a further advantage in that it would provide competing centers of analysis within the Agency on a single discipline competition which is essential to the development of differing interpretations on critical intelligence issues (especially on issues where expertise is lacking elsewhere in the community).
There are, of course, problems with this approach that involve: defining the sometimes grey area between current and in-depth research, the potential ivory-tower dimension of a research shop divorced from the "juice" of current events, and the personnel costs of two shops on the same geographic areas using essentially the same disciplines. Still, the alternative of trying to meld both needed tasks in one staff or office would appear to be an almost sure formula for further slippage in the Agency's vital research capability. In order to keep the research components from becoming too far removed from the mainstream of current affairs, they could be given responsibility for estimative-type intelligence, which requires continuing attention to the flow of current events. The goal would be to protect the Agency's research capability from current demands while at the same time preserving the access of current intelligence components to the in-depth knowledge so essential to their reporting.
The Interdisciplinary Problem12
As more intelligence issues become interdisciplinary and transnational in nature, CIA will have to strengthen its ability to produce such studies. The Agency is today the only producer of finished intelligence in the community that has the analytical resources to turn out interdisciplinary studies — and has done so successfully — yet it is organized along disciplinary lines which do not readily encourage the production of such studies. One obvious solution would be to combine the disciplines along geographic lines in a new organizational structure. This would foster interdisciplinary approaches to national and regional problems. At the same time, however, it might discourage the intensification of expertise that is provided within a single disciplinary framework and might not, in the end, produce an improved product. Depth and continuity of expertise is a key CIA strength. Although some way of overcoming the bureaucratic obstacles and disincentives of the present system is needed, extensive changes in formal organizational structure are probably not the answer to the interdisciplinary problem.
No organization in or out of government appears to have tackled the interdisciplinary issue with fully satisfactory results as yet. The most successful efforts seem to have involved ad hoc as well as semipermanent teams; but there are mixed views within CIA on the value of team approaches. In applying team approaches to interdisciplinary problems in the intelligence context, consideration of the following factors may increase the chances for success:
— the personnel must be carefully chosen in each instance from the standpoint of ability to work in a group context; they must also combine real expertise with serious intellectual curiosity and interest in the other disciplines;
— the home components must be bureaucratically willing to concede the necessary disciplinary sovereignty to the collective judgment of the team and its leader;
— the officers involved must believe the team effort is a vital part of their professional duties for which they will receive full professional recognition;
— there must be a real effort at an exchange of ideas and approaches by various disciplines throughout the project aimed at the development of a true synthesis, rather than the production of component sections in isolation, followed by a cut-and-splice job by the team leader at the end of the effort;
— team leaders must understand the techniques of team management; this has a training dimension largely untouched as yet in the CIA.
Some of the Agency's team efforts have accomplished some of these requirements; others have foundered because they were lacking. But insistence on the application of such criteria by management in interdisciplinary undertakings should result in immediate, practical, and inexpensive improvement in this type of product.
Consideration should be given to other interdisciplinary approaches as well. In particular, analyst training, carefully developed for its relevance to intelligence concerns, should be undertaken to broaden analyst appreciation and understanding of interdisciplinary problems. This probably could be done most effectively through inside training and would take extensive and sophisticated development of pertinent training curricula. Recent OPR efforts to provide internal multidisciplinary training for its analysts are a step in this direction.
In order to provide a more permanent organizational mechanism for stimulating and improving the Agency's interdisciplinary work, broad-gauged senior officers could be appointed to monitor DDI production with a view to suggesting ways in which interdisciplinary angles could be incorporated into and synthesized in all appropriate products. Such individuals would, as generalists, supplement NIO monitoring efforts. The NIOs are organized along set geographic and disciplinary lines. A main question would be whether this new function could best be performed within the confines of the existing office structure — with an interdisciplinary officer for each production office — or whether it could most effectively be accomplished outside the present structure — by a full-time coordinator for interdisciplinary studies at the Directorate level. Perhaps a two-pronged approach is needed. In any case, the most essential ingredient will be a positive approach to the interdisciplinary problem on the part of production office management.
One additional means of extending CIA's own resources and of bringing a greater range of assets to bear on the future intelligence environment predicted in this study can be accomplished in theory by greater dependence on inputs from outside the government, principally from academia and private industry, in our finished intelligence undertakings. Our successes so far in this sphere have been infrequent, except in the scientific and technical fields where there is a long tradition of outside participation which has contributed importantly in many instances to the product. In the area of social science, our experience in using outsiders seems to be improving lately as we recognize that the key to success is the closest possible continuing linkage between our staff and the outside experts, so that the objectives, relevance, and information central to the undertaking are clearly understood and repeatedly scrutinized by both elements. This means extra managerial time, but the pressures on us to utilize outsiders are unlikely to diminish and, in the social sciences, we may find that as we develop a cadre frequently utilized and closely tuned to our substantive interests, their value in increasing both the quality and the range of our finished intelligence product will be significant.
Brown, Seyom, New Forces in World Politics (Brookings Institution, Washington, DC, 1974).
Center for the Study of Intelligence, CIA Intelligence Support for Foreign and National Security Policy Making, SECRET (CIA/OTR, January 1976).
Commission on the Organization of the Government for the Conduct of Foreign Policy, Volume 7 on Intelligence Functions (U.S. Government Printing Office, Washington, DC, 1975).
Director of Central Intelligence, Perspectives for Intelligence, 1976-1981, SECRET (USIB/IRAC, October 1975).
International Research Council, Foreign Policy Contingencies: The Next Five Years (Center for Strategic and International Studies, Georgetown University, Washington, DC, 1975).
Kahn, Herman (project director), The World, 1982-1991 (Hudson Institute, Groton-on-Hudson, New York, 1972), prepared for DIA, Contract #DAHC15-72-0131.
Overholt, William H. (project leader), Forecast for 1994 (Hudson Institute, Groton-on-Hudson, New York, 1974), prepared for DIA, Contract #DAAB09-74-C-0014.
Owen, Henry (ed. ), The Next Phase in Foreign Policy (Brookings Institution, Washington, DC, 1973).
Proceedings of National Security Affairs Conference, 8-10 July 1974, Defense Planning for the 1980s and the Changing International Environment (National War College, Washington, DC, February 1975).
Senate Select Committee to Study Governmental Operations with Respect to Intelligence Activities, Foreign and Military Intelligence (U.S. Government Printing Office, Washington, DC, 1976), Book 1.
Some Likely Key Intelligence Questions for the 1980s, SECRET (CIA/ORD, June 1974).
The Civil Technology Assessment Program: A History and Evaluation, CONFIDENTIAL (CIA/OSI, April 1976).
1 This is adapted from an intelligence monograph originally issued by the Center for the Study of Intelligence. Ross Cowey was the principal author with contributions and assistance from the staff and fellows of the Center.
2 It should be noted that the forecasts in this paper were made in August 1976 without knowledge of the outcome of the Presidential elections or of reorganizations proposed or implemented within CIA since August.
3 OCI also took over production of the political sections of the National Intelligence Surveys in 1962, after State withdrew from the NIS program for lack of resources; in 1974, the entire NIS effort was abandoned because of the combination of diminished resources and diminished demand.
4 This expansion of CIA's market was recognized and institutionalized by Executive Order 11905 of February 1976, which provides that CIA shall "produce and disseminate foreign intelligence relating to the national security, including foreign political, economic, scientific, technical, military, sociological, and geographic intelligence, to meet the needs of the President, the National Security Council, and other elements of the United States Government" (emphasis added). The Departments of Treasury, Commerce, Agriculture, and other governmental components not normally or originally part of the national security arena, now have become regular consumers of finished intelligence.
6 The investigations themselves, of course, led to a dramatic increase in the number of congressional requests for information from CIA. In 1975, Agency officials appeared before congressional committees some 90 times — nearly twice as often as in the previous year, and more than five times as often as in 1971 and 1972. Only substantive intelligence and briefings have been considered for the purposes of this report, but even here the increase has been considerable, and much of the support today is being provided on a routine basis (e.g., the National Intelligence Daily and regular committee briefings).
7 It should also be kept in mind that the more our product is exposed to Congress, the more the opportunity and perhaps the impulse for Congress to carp about our"mistakes" and "failures," all of which takes time and assets to rebut.
10 Interestingly, as the intelligence market has expanded in recent years, the number of longer finished intelligence reports or monographs produced by CIA has declined slowly but steadily, while the number of short, current pieces has increased significantly. The net result is that more finished intelligence items now are being produced on a wider variety of subjects than heretofore, with approximately the same analytical and production resources.
11 It is generally recognized that components which deal with long-range problems or planning should be organizationally distant from responsibility for current operations; the same would be true for components doing in-depth research. See Anthony Downs' Inside Bureaucracy (Little, Brown & Co., Boston, 1967) for a more detailed discussion of this and other relevant organizational rules or hypotheses.