Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
APPENDIX TWENTY DILEMMAS OF OIL VULNERABILITY Robert Axelrod The United States imports almost 20 percent of its total energy and an even higher percentage of its oil. If a substantial part of this supply were cut off, the American economy and society would suffer major disruptions, to say the least. This much is obvious and a proper cause for concern. What is not so obvious is that oil vulnerability results in many dilemmas. The effort to solve one part of the problem may only make another part worse. The perspective of this paper is to regard the United States as a huge, loosely coupled adaptive system, and a potential oil shortfall as a crisis that gives a shock to the system. Adapting to a sudden, novel, and very large shock is not something most systems can do gracefully. To anticipate the potential pathologies of the adaptation process may help us alleviate some of the worse effects. This paper describes a series of dilemmas under three headings: those that arise in preparing for a crisis, those that arise from an interaction between precrisis planning and crisis response, and those that occur only after the onset of the crisis. For their helpful suggestions I thank Elliot Aronson, James March, Amy Saldinger, Paul Stern, and Tom Wilbanks. This work was supported by grant no. SES 80-23556 from the National Science Foundation. For a recent statement of the problem and suggestions for coping with it, see U.S. Department of Energy (DOE) (1980b). 55
56 DILEMMAS OF PREPARATION One of the best ways of planning for a shortfall of imported oil is to build up a reserve stockpile. This can alleviate the severity of any shortfall, especially in the early stages of a crisis before much adaptation can occur. By making the United States less vulnerable, a reserve serves two other purposes as well: it helps deter a boycott by showing that the United States can better resist political blackmail, and it helps reduce the pressure for military moves to guarantee imports from a troubled area. Almost everyone who has studied the energy crisis agrees that building up a substantial strategic petroleum reserve is one of the best policies the United States can undertake.2 But the United States has fallen far short of its target for building up a strategic petroleum reserve, having achieved only one-fifth of its target for early 1981 (Martin, 1981). If a petroleum reserve is so helpful, why hasn't it been achieved? The answers to this question encompass the first set of dilemmas of oil vulnerability. 1. The Procrastination Dilemma: Now is never a good time to add to an oil stockpile. When the economy is doing poorly, with high inflation and high unemployment, budget reductions are the order of the day. When the economy is doing well, there is a strong demand for oil and purchases for a reserve would drive the prices even higher. Another reason that the United States has not been building up the strategic petroleum reserve is that Saudi Arabia warned us not to (Stobaugh and Yergin, 1979, p. 52).3 They would prefer to keep the United States vulnerable and are willing to threaten reductions in their own production to maintain U.S. vulnerability. 2See Stobaugh and Yergin (1979:52-53), Schurr et al. (1979:430-435), U.S. President's Commission for a National Agenda (1980:48), DOE (1980b:29-35), Energy Policy Task Force (Halbouty Report) (1980:14-15), Heritage Foundation (1980:94-95), Alm (1981), and DOE (1981). 3 The purchases stopped in August 1979 according to DOE (1980b:30). The purchases began again in late 1980 or early 1981, according to Martin (1981).
57 2. The Democratic Dilemma: The more democratic the politics of energy preparations, the fewer the preparations are likely to be. Democracy's strong suit is the passage of programs that give benefits now and have costs later. Stockpiling does just the reverse. An energy emergency may not come in the term of the current president, nor in the terms of many current members of Congress. It is a problem that can be passed on to the next guy. Even without economic and foreign policy reasons to procrastinate, the political process is biased against costly preparations for uncertain crises. 3. The Public Interest Dilemma: The more widespread the potential harm of oil vulnerability, the less focused are the pressures to deal with it in advance. Special interests are the traditional means to mobilize support for effective action in a democratic political system. But since the potential harm caused by oil vulnerability is so widespread, it is difficult to focus attention and political resources on the problem. Everyone's problem is no one's problem. 4. The Intimidation Dilemma: The more it disturbs exporters to have us building a reserve, the more valuable the reserve is. The very fact that exporters are concerned about the U.S. stockpile is an indication that they take seriously the possibility of a boycott or major price increase. Giving in to threats rather than building up the stockpile is admitting vulnerability and guaranteeing that it will continue. 5. The Apprehension Dilemma: The less we fear a boycott, the more likely it is. A boycott is not like an earthquake. The probability of a boycott occurring is in part dependent on our own actions. The less we prepare for it, the more vulnerable we are, and hence the more attractive it becomes as a policy tool of an exporter or group of exporters. The less well prepared we and other importers are, the less a sacrifice of revenues will be needed by exporters in a contest of wills and the more tempted they may be to initiate such a contest when serious disagreements arise. An important implication of this point is that we should calculate the value of a reserve not only in terms of what it can achieve if a major shortfall should occur, but also in terms of the reduced likelihood that such a problem will ever actually arise. 6. The Preparations Dilemma: Any major governmental program to prepare for an oil shortage deters decentralized adaptation. A large stockpile of oil, for
58 example, may lead individuals, businesses, and local governments to do less on their own to prepare. To the extent that federal preparations are seen as successful, others will feel that there is less need for them to bear additional costs of preparation. 7. The Learning Dilemma: Learning from experience may be maladaptive. Experience may n teach that preparations are wasteful. From the point of view of an industrial firm, investments in crisis preparations will lower profits in the period before a crisis. This can hurt a firm relative to its less prudent competition. If there is a small crisis, everyone may notice that the expensive preparations were in fact not a profitable investment, and the conclusion based on this history might well be that expensive precrisis adaptations are simply not warranted. The same conclusion may be learned by consumers who notice that their painful adaptations to lessen their energy use have not yet paid off. Thus a record of false alarms and minicrises may be effective in drawing attention to the problem but may also lead to maladaptive behavior in terms of the potential for a large crisis. (For another facet of the learning problem, see dilemma 10 below, the warning dilemma.) DILEMMAS OF ANTICIPATION Measures anticipated to be taken during an energy emergency influence individual decisions taken before an emergency. Prior adaptation to anticipated conditions of an acute shortage may not be functional for adapting to a chronic shortage. And vice versa. 8. The Efficiency Dilemma: Energy efficiency reduces emergency flexibility. To the extent that homes, automobiles, and factories become energy efficient in response to the chronic energy problem, there is less flexibility to achieve further quick gains when an acute shortage occurs. What is true for hardware is also true for behavior. For example, to the extent that commuters use car pooling to adapt to high energy prices, there is less slack in the system for further adaptation in the face of an emergency. This problem will be alleviated if precrisis adaptations make people more aware of the extent to which they can adapt even more should a crisis occur. Of course, energy efficiency reduces energy use before the crisis ever begins and this helps reduce the need for oil imports in the first place. And reduced energy use
Be will make a given stockpile last longer. Nevertheless, to the extent that the entire system adjusts to this lower level of demand, it is inevitable that energy efficiency can reduce emergency flexibility. 9 The Equity Dilemma: Decentralized efforts to . prepare for a crisis will be deterred by an expectation of effective demands for equity during a crisis. Why should a firm stockpile its own energy supply if it expects that its stockpile will be confiscated when an emergency actually arises? The expectation that precrisis notions of fairness will be imposed by fiat can seriously reduce the incentives of an individual, an industry, or a local government to make sacrifices now for benefits that it may be forced to share later. For example, why should a city or school board be the only entity in a county to be prepared if the costly preparations would have to be diluted with the profligate policies of neighboring towns or school districts? The equity dilemma is a fundamental one that arises from the very real conflict between achieving a fair distribution during a crisis and lessening the overall shortfall should one occur. The very expectation that the fairness will be achieved makes it easier for everyone to let the other fellow prepare. It applies to all types of decentralized preparations that are subject to demands for sharing. To the extent that there is a politically viable "right to energy, it prevents effective decentral- ized preparations, since even those who do not prepare will then be able to claim an equal share in access to scarce resources. The equity dilemma arises in its worst form if decentralized and sharable preparations are deterred by expectations of postcrisis demands for sharing. Unfortunately, this may be the most common case. Many preparations besides stockpiling are sharable. For example, if one city purchases extra fare boxes to prepare its school buses to be used as commuter buses in an emergency, it may be required to lend some of these fare boxes to a neighboring city in an emergency. Decentralized preparations can be the most effective type for many purposes. And unfortunately, a centralized response may be the most common one for the implementation of demands for equity. (See also dilemma 19, the centralization dilemma, below.) The equity dilemma can be alleviated in several ways. First, the preparations can be centralized, as in a national petroleum reserve. Then there is a collective
60 cost for making the preparations, and a collective benefit in their equitable distribution. A way of reducing the equity dilemma but keeping decentralized incentives would be some form of credible guarantee that whoever undertook the costs of the decentralized preparations would be allowed to keep a large share of the benefits. This could be done by the federal government by a variety of means. One would be dismantling the federal gas rationing apparatus to make it less likely that price controls would be imposed in an emergency. Another method of achieving credibility that benefits could be retained would be to clarify in advance that decentralized preparations are in the national interest, that everyone has fair warning that an emergency may come, and that those who prepare should be able to benefit (at least in part) from their foresight, which would serve us all. But the poor can not prepare very well. Rationing and/or price controls may be the only effective ways of achieving fairness for them in an emergency. Hence the equity dilemma. 1O. The Shared Cutback Dilemma: steps taken before a crisis will be deterred by an expectation of demands for proportional cutbacks during a crisis. Once a crisis begins there will be demands for everyone to cut back on energy usage. These demands may well take the form of percentage cuts based on usage before the crisis. 4 The possibility of a percentage cutback can reduce the consumers' incentive to make reductions before the crisis begins. Of course, there are emergency measures that can avoid this problem, such as rationing gasoline with a white market in coupons. But for uses like home beating and industrial activities where "fair n consumption varies greatly from consumer to consumer, percentage cutbacks may seem logical in a crisis. And in these cases, the shared cutback dilemma applies with full force to reduce incentives to be efficient in advance. 11. The Warning Dilemma: Vigilance breeds fatigue. The more emphasis there is on providing warning of the onset of a major crisis, the more likely that there will Energy efficiency 4A good case study based on the experience of electricity rationing in Los Angeles in 1973-1974 is Acton et al. (1974) and Acton and Mowill (1976).
61 be false alarms. And many false alarms will lead to a discrediting of the warning mechanism. This is the familiar problem of the boy who cried wolf. Or of the home smoke alarm that was not heeded because it often was set off by cigarette smoke. Or the intelligence official who correctly predicted 19 of the last 2 wars. The problem is a fundamental one, well described by signal detection theory (Coombs et al., 1970:165-201). The implication is that a warning of an uncertain emergency is costly because false alarms eventually retard effective adaptation and discredit the warning system.s 12. The Inventory Dilemma: Unfilled end-use storage capacity is destabilizing. When private and commercial users of energy have unused capacity, they are likely to buy more fuel when shortages or major price increases appear imminent. This leads to a self-fulfilling prophecy, as the distribution system draws down its stocks in response to the increased demand. This self- fulfilling prophecy can destabilize the system, even in the absence of a shortfall of domestic production or imports. Thus for end-users to purchase storage capacity in anticipation of a crisis can be destabilizing if the storage capacity is not kept nearly full. The problem would be amplified if the federal government undertook a crash program to build up the strategic petroleum reserve when a crisis appeared imminent. This brings out the contrast between storage capacity and actual stockpiles. Stockpiles are stabilizing, but unfilled end-use storage capacity can be destabilizing. Another implication is there is a big difference in these terms between a crisis that comes on slowly and allows storage capacity to be filled gradually and a crisis that comes on unexpectedly. Still another implication is that a sudden threat of a crisis can lead to wild oscillations in supply and demand. After the end-users fill their storage capacities, they may draw down on them if the crisis does not materialize as expected. This will lead to lower demand from the distributors who may then cut back on their order, and by the time the end-users are back to normal, the distributors may be low. This oscillation can extend right back to producers, giving "boom and bust" demand signals, and resulting in highly erratic sFor useful suggestions on minimizing the false alarm effect, see Breznitz (1976).
62 responses to highly erratic signals. 6 If this instability can occur in the context of a genuine import shortfall the early inventory oscillations driven by increased demand may be confused with the shortages driven by falling supply. The demand side may cause a run-up and collapse of prices before the import shortfall is directly felt. If so, then the falling prices may be misinterpreted as indicating that the supply shortage was a false alarm and that painful emergency measures are unwarranted. Thus the inventory dilemma may lead to the undermining of effective emergency adaptation just when it is needed most. Moreover, if the oscillations occur in the absense of a subsequent shortfall, then the whole episode will serve as a false alarm that can undermine future adaptation, as spelled out in dilemma 11, the warning dilemma, above. DILEMMAS OF CRISIS RESPONSE While some dilemmas appear in the preparation stage, and others occur as a result of the anticipation of actions to be taken should a crisis occur, there is a third category of dilemmas that arise only after an emergency actually occurs. 13. The Protection Dilemma: The more categories of energy users that are Held harmless" by a crisis response, the more the others suffer. Take a small example. The gas rationing plan of the Carter administration exempted a list of users, including car rental firms, from a percentage reduction in gasoline supplies (DOE, 1980a). This left less for others if there were a serious shortfall. The effects are twofold. First, it means that the berm done to others is greater because they will have to take more drastic measures to adapt. Second, it undermines the feeling of equal suffering, which is so necessary for voluntary compliance with other regulations. These two effects occur whether the exemption is a category of user (e.g., rental car companies using gasoline) or a category of energy (e.g., home heating oil and diesel oil). The attempt to put the burden of the entire shortage on one 6For a good treatment of instabilities caused by inventory oscillations, see Forrester (1961).
J 63 part of the refined product, gasoline, leads to the same protection dilemma. 14. The Alliance Dilemma: Every consuming nation is tempted to protect itself through bilateral deals with supplier countries, but this would hurt all the consumers. This is an example of the famous e-person Prisoner's Dilemma (Olson, 1965). If everyone pursues his or her own private interests, everyone will suffer. Another oil shortfall could panic Western alliance members into making their own deals with producing nations, and these deals might involve political and economic concessions that would be highly damaging to the alliance as a whole (Alm, 1981). 15. The Optimism Dilemma: The more an embargo is hurting the United States, the more the President will be tempted to say it isn't. There are two reasons. First, there is the short-term desire to get domestic holders of supplies to stop hoarding them. This may be especially true if the cause of the problem appears to be the inventory dilemma (see dilemma 11 above). Second and more important, there will be a need to put on a good face for those who are attempting to pressure the United States by means of a boycott. The president will want to make this effort appear futile and hence will want to appear optimistic. But in an extended crisis, optimism is maladaptive because it deters necessary adjustments and because it will soon discredit the optimist. An implication of this dilemma is that during a boycott the public should rely on unofficial information about the seriousness of the problem rather than giving full credit to what the federal government says. Multiple channels of information and prediction are likely to be useful supplements to official pronouncements in any case. Among the numerous channels of private information are automobile club reports on gasoline availability, opinions of energy and foreign policy specialists, and investigative reporting by the mass media. 16. The Reassurance Dilemma: The more the public believes government assurances that there is no need to panic, the less adaptation there will be. Ugly problems in the gas lines are just one reason why the government will want to reassure people that the situation really is not so bad and is fully under control. A second reason is the self-fulfilling prophecy of the inventory dilemma (dilemma 11, above). Still another reason is the need to put up a good front in the case of a politically motivated boycott, as expressed in the optimism dilemma (dilemma 15,
64 above). Whatever the reason, reassurance that the situation is not so bad after all will tend to deter adaptation, especially medium- and long-term adaptation. 17. The Cake Dilemma: You can't eat into Your strategic reserve and have it, too. There are two very different and competing uses of a strategic reserve in a crisis. One use is to alleviate the worst of the problems by drawing on the stockpile in the period before individuals and institutions have a chance to adapt to the new situation. Many of the most effective adaptations will take weeks to implement. These include, for example, setting up car pools, changing school hours so that school buses can serve on commuter routes, and setting up gas allocation operations. In the meantime, the shortages might be so severe that many plants would have to close down unless emergency supplies were available. But if stockpiled fuel were used at the start of a major shortfall, then the government would be largely unable to respond to latter problems as they arose. This could leave the nation feeling naked. As bad as things may be at the beginning of a crisis, it might be useful to have a collective resource to provide a safety net in case things got even worse. This would be especially true if the shortfall were caused by a boycott, since then a stockpile-in-being would demonstrate that we were not about to be cowed. 18. The Planning Dilemma: In a crisis, plans seem worthless, but actions are limited to capabilities actually available. Leaders do not wish to follow blindly contingency plans adopted by others with little knowledge of the context of the actual crisis and little appreciation of the political problems of the leaders. Thus plans seem worthless and are often ignored in a crisis. But leaders are in fact constrained by the capabilities that are actually available. For example, opting to use the strategic petroleum reserve to alleviate the worst of a crisis in the first few weeks will require a very large pumping capacity. Without that pumping capacity there will not even be a choice. AS another example, ration coupons cannot be used immediately if they have not been printed in advance. Thus the functional role of plans is much different from the nominal role. The functional role of plans should be to identify bottlenecks in the adaptation process so that capabilities will exist at the time of the crisis to avoid these bottlenecks if desired. This is quite different from the nominal role of plans as drafts of
65 policies to be signed at the time of a crisis. If planners misunderstand their role, then leaders in a crisis will be faced with the dilemma of choosing among sterile contingency options or having to adapt without the capability of doing so effectively. 19. The Centralization Dilemma: In an emergence there are likely to be demands for greater centralization of decision making, but these demands will almost certainly be mistaken. Centralization gives the illusion of control and efficiency. Many will insist that the federal government "do something n and undertake Coordinated actions and end ~inconsistencies. n Yet, for rapid adaptation, greater decentralization of decision making is likely to be more effective. 20. The Trust in Government Dilemma: The more that trust in government is needed, the less there will be. If a crisis comes without adequate preparation, the public will angrily ask why the government did not make timely preparations. To adapt effectively to a crisis when there are inadequate preparations, however, will probably require a good deal of trust in the validity of the description of the events that the government is offering, and in the recommendations for painful action it is proposing. Yet, the less adequate the preparations have been and the more painful the adaptations have to be, the less reliance the public is likely to have in the ability of the government to handle the crisis for which it did not prepare. CONCLUSION Viewing the United States as a huge, loosely coupled adaptive system has revealed many more interaction effects than would be suggested by a more simple organizational or market model. And, as we have seen. many of these interactions lead to dilemmas. Understanding the dilemmas of oil vulnerability will not make the dilemmas go away, but it can help us to appreciate better the consequences of the choices. REFERENCES Acton, Jan Paul, Graubard, M. H., and Weinscbrott, D. J. (1974) Electricity Conservation Measures in the Commercial Sector: The Los Angeles Experience. RAND
r 66 R-1592. Unpublished report prepared by the Rand Corp. for the Federal Energy Administration. Ac ton, Jan Paul, and Mowill,Ragnhild (1976) Regulatory Rationing of Electricity Under a Supply Curtailment. RAND P-5624. Unpublished report prepared by the Rand Corp. Alm, Alvin L. (1981) Energy supply interruptions and national security. Science 211:1379-1385. Breznitz, Shomo (1976) False alarms: the effect on fear and adjustment. Pp. 127-140 in Irwin G. Sarason and Charles D. Spielberger, eds., Stress and Anxiety. Volume 3. New York: Wiley. Coombs, Clyde H., Dawes, Robyn, and Tversky, Amos (1970) Mathematical Psychology. Englewood Cliffs, N.J.: Prentice-Hall. Energy Policy Task Force (1980) Report. (Halbouty Report), November 5. Forrester, Jay (1961) Industrial Dynamics. Cambridge, Mass.: MIT Press. Heritage Foundation (1980) Project Team Report for the Department of Energy. First draft, October 28. Martin, Douglas (1981) "Store Oil" is easier said than done. New York Times, February 15. Olson, Mancur, Jr. (1965) The Lcgic of Collective Action. Cambridge, Mass.: Harvard University Press. Schurr, Sam, et al. (1979) Enemy in America's Future (A Resources for the Future Study). Baltimore, Md.: Johns Hopkins University Press. Stobaugh, Robert, and Yergin, Daniel, eds. (1979) Energy Future: Report of the Energy Project of the Harvard Business School. New York: Ballantine. U.S. Department of Energy (1980a) Progress Report to Congress on the Standby Motor Fuel Rationing Plans. DOE/RG-0038, June. U.S. Department of Energy (1980b) Reducing U.S. Oil Vulnerability: Energy Policies for the 1980's. DOE/PE-0021, November 10. U.S. Deparment of Energy (1981) Public Discussion Package of the Third National Energy Plan, DOE/PE-0022, March. U.S. President's Commission for a National Agenda (1980) A National Agenda for the Eighties. Washington, D.C.: U.S. Gove r Dent Printing Office.
