|
McNair Paper 63, All Possible
Wars? Towards a Consensus View of the Future Security Environment,
2001-2025, November 2000
Estimates, Forecasts, and Scenarios
People have an innate ability to build scenarios, and to foresee the future. --Peter Schwartz30
Estimates Estimates utilize an assessment of current conditions to identify possible future events. This method is most closely associated with official intelligence estimates provided by intelligence agencies and services, the most significant of which are the National Intelligence Estimates (NIEs) summarizing assessments common to the overall intelligence community. Such intelligence estimates generally combine current information on a variety of elements--such as industrial production, technology trends, and military orders-of-battle--in a manner that is comprehensive enough to identify probable near-term policies and events. Due to Cold War controversies, as well as natural conservatism and bureaucratic pressures for continuous accuracy, most official intelligence estimates focus almost exclusively on capabilities of potential opponents and shy away from discussion of likely intentions.31 But whether including intentions or not, the priority remains accuracy, which requires a relatively short time horizon. Department of Defense net assessments generally fall under the category of estimates. Forecasts Forecasts represent longer-range assessments, primarily relying on trends-based analysis. Most credible forecasts are issue-specific, generally under the assumption that an issue-area expert is best qualified for making an assessment concerning the continuity or modification of current trends. When issue-oriented forecasts are combined in an attempt at comprehensiveness, variations of the Delphi Method--in which experts are polled as to their views--appear most often used.32 Although most future assessments produced today can best be considered forecasts, the term is frequently disparaged by futurologists of the burgeoning "futures industry" who favor the use of scenarios. As one source admits, "the success of forecasting is decidedly mixed, especially so in industries that are experiencing discontinuous change....Forecasting...has a long history of unreliability when it was wrongly used to predict the unpredictable."33 However, the same compendium advises: "We would suggest that organizations need to employ both technologies [forecasts and scenarios], because forecasting does shed light on how predictable trends may combine to produce significant changes in the business environment."34 Forecasts, along with the futurologists themselves, are subject to considerable criticism from policy analysts. As the late Harry G. Summers, a prolific author and retired U.S. Army colonel, argues: Although futurologists like Alvin and Heidi Toffler make their livings in claiming to predict coming events, their 1993 effort, War and Anti-War: Survival At the Dawn of the 21st Century, like other such works, is at best an exercise in scientific wild-ass guessing. Unless taken to heart and acted upon, most such attempts are harmless, and may even offer some minor insights. But the future is and will remain uncertain.35 Ironically, forecasts can be implicit, and as such, appear in almost every analytical work on future policy. This includes the very work in which Colonel Summers dismisses the Toffler forecasts, which is subtitled "A Military Policy for America's Future."36 Since forecasts are not necessarily explicitly labeled as such, and appear at least implicitly in every strategic assessment, a first step in evaluating the validity of any policy recommendation is to determine the assumptions about the future, i.e., the forecast, on which the recommendations are based. This is a preliminary step that is not always followed in debates on defense policy. Scenarios Scenarios can be thought of as a range of forecasts, but both their construct and intent are more complex. In defense analysis, scenarios can be traced back at least to Herman Kahn's Thinking About the Unthinkable approach to analyzing potential nuclear wars that might occur if deterrence failed.37 The current popularity of scenarios in business planning is largely the result of Pierre Wack's strategic business planning for Royal Dutch/Shell. Wack is often credited as the sole forecaster of the rise of OPEC and the oil crisis of the 1970s; however, scenario builders are quick to point out that their objective is not to forecast a particular future at all, but to help "to make strategic decisions that will be sound for all possible futures."38 In the words of Wack's collaborator, Peter Schwartz, who had a significant role popularizing scenarios work in the United States, the breakthrough in scenario development came about when Wack changed from "developing simple tales of possible futures" to building descriptions of "full ramifications" designed to "change our managers' view of reality."39 Thus, modern scenarios tend to be richly developed depictions of alternate worlds based on plausible changes in current trends. "The end result, however, is not an accurate picture of tomorrow, but better decisions about the future."40 This is the significant difference between scenarios and forecasting; presumably, forecasts are attempts at an accurate, ostensibly predictive picture of the future. The technique of scenario building has become professionally formularized. Usually done with groups of diverse subject matter experts, the initial step is determined by the drivers that will propel future change. Drivers are the underlying factors in current trends, such as population growth or decline, technological development and diffusion, or human factors like the will to power. Changes in drivers result in changes in trends, which, in turn, result in changes in the human environment. A scenario is a depiction of the future based on the selected directions of a series of drivers. Because of the multiple directions possible for multiple drivers, the number of scenarios required to depict all plausible future outcomes can be rather large. The heuristic effect of considering the difference in implications of the multiple plausible future outcomes provides for a strategic conversation that allows decisionmakers to consider implications that may not be evident in the reality of today.41 The differing implications of multiple scenarios thus provide for a wide range of policy options to analyze. Like theories, and unlike forecasts, scenarios are neither right nor wrong, merely plausible or implausible. Despite the quotation opening this chapter, the innate ability developed through scenarios is not to foresee the future, but a range of possible futures. Scenario work is used increasingly by defense planners because the development of a range of alternatives corresponds well with the traditional military planning process of anticipating all possible moves of enemy forces. The would-be Napoleons of history rarely considered only one possible move or one possible response. Of the military services, the Air Force has placed the greatest resources toward formal futures scenario development, with a significant effort culminating in late 1996.42 Project 2025, a study conducted by Air University for the Air Force Chief of Staff, developed eight alternative world futures and conducted an analysis of the defense policy implications of the four assessed as "providing the most stressful planning challenges."43 Other service efforts have generally focused on two or three alternative worlds, or on specific technological trends. Methodology
Comparing the strengths and weaknesses of the three primary methodologies for futures assessment reveals implications for policy recommendations. The strengths and weaknesses of the many competing defense policy recommendations are themselves influenced by whether their expectations are derived from near-term estimates, longer-range forecasts, or insights from scenario building. Theoretically, the time frame for which the policy recommendation is intended would dictate the method or mix of methods utilized. However, rarely are the methods used clearly and distinctly identified.
|
|
|
|
As summarized in the table above, estimates have the strength of a greater degree of definition that appears directly applicable to practical, relatively near-term decisionmaking. But the reliance on accuracy in an environment with multiple variables mandates the examination of a relatively short time frame of events. Political and technological trends often do not proceed in a linear manner, and therefore defy prediction over a long period.44 Defense policy recommendations based on estimates may assuage immediate concerns but may not capture the range of possible long-term concerns against which a prudent planner might hedge.45 In contrast, forecasts capture a longer time frame, but their ultimate accuracy is subject to events that cannot be predicted with certainty. Many forecasts make up for this vulnerability by examining a very specific topic or small slice of possible futures. Presumably, the narrower the topic, the more specific--and therefore the more accurate--the forecast. Unlike scenario building, forecasting need not take a holistic approach toward the future. For example, forecasts are routinely made on the future profitability of a particular corporation or industry. Indeed, most of the decisions made on Wall Street or in commodity futures trading are based on forecasts with much the same characteristics as the most outlandish writings of futurists.46 And like the plethora of conflicting financial advice, there is considerable contradiction between forecasts. The validity of forecasts is assumed to correspond to the expertise of the forecasters themselves. To get the best forecast, the common approach is to find the most experienced or credentialed expert. Indeed, forecasting encourages the creativity of subject matter experts, requiring them to go beyond the safer realm of estimates. The element of creativity promotes the comparison of diverse viewpoints, and many forecasts are compiled by committee in order to ensure all possibilities are considered and analyzed. This simplifies planning and makes the forecast a more acceptable tool for decisionmakers used to relying on the collective wisdom of their staffs. However, the existence of contradictory forecasts creates an insidious tendency toward extreme forecasting. Outrageous statements are often made in order to attract attention to otherwise responsible forecasts, as often by media reports as by the forecasters themselves. There is an even greater tendency to claim an unjustified degree of certainty. Scenarios have a heuristic orientation, and thus do not need to demonstrate an accuracy for prediction. The intent is to be inclusive of all possibilities, even contrarian thinking. In order to discourage the perception that scenarios should be predictive, Pierre Wack referred to scenarios as "the gentle art of reperceiving."47 "Reperceiving" consists of questioning assumptions about the world.48 Peter Schwartz advises the use of "remarkable people," unconventional thinkers "found in unconventional locations and roles" to ensure the development of inclusive scenarios.49 Freedom from the need for direct prediction promotes a longer-range look at alternative futures and allows for the development of hedge strategies toward unlikely, but possible events. However, the heuristic approach requires a methodology for translating insight into practical policies. This translation process often requires more intellectual effort than the process of scenario-building itself. Likewise, it does not necessarily lend itself to immediate, problem solving decisions. The need for translation makes scenarios less attractive to practical decisionmakers, who are likely to view scenario efforts in the same light as Summers views forecasting by the Tofflers: harmless, and even offering some minor insights. But the process of scenario building lends itself to conferences, workshops, off-sites, and other methods of modern management, thereby ensuring its popularity as an appropriate public demonstration of thinking about the future. Though based primarily on estimates and forecasts, both the National Defense Panel report of December 1997 and the U.S. Commission on National Security/21st Century's New World Coming include brief chapters identifying four potential future scenarios.50
The inevitable question as to which is the best methodology has a simple answer: it depends on the desired balance between certainty and insight. If time and resources permit, an examination including estimates, forecasts, and scenarios would prove the most comprehensive of crystal balls. The sources selected for this study represent exactly that sort of mix.
|