ANNEX A

The E Process and National/Local Aviation Development: Some Examples

In our overview paper on the E process and Aviation development, we described generally how the process could be used to overcome public goods and externatity problems affecting aviation development. This paper elaorates on the method in the context of a general method for resolving general externatity problms in an urban/regional development setting. It will illustrate first, a process for determining national budgetary expenditures for aviation development such as expenditure on the national air traffic control system (this is our concrete analogue to the pubic goods expenditure problem in the overview paper). The paper then explores a solution to the National aviation noise problem (the solution to a perceived public bad from negative externality effects associated with aviation development. Finally, we illustrate a means by which communities could encourage the privatization of aviation development at the State and local level, consistent with national aviation development goals and local preferences for sound economic development and realization of a broad array of environmental and social goals.

To fully understand the examples presented in this annex, one must appreciate the broad social goals that are represented in any attempt to implant the E - procedure in a system of local decision-making and translate these decisions or preferences into national-level decisions. For this reason, the annex is meant to be read in conjunction with the annexes which accompany it, which explain the assumptions underlying incentive compatibility as a system of arriving at more effective social valuations -- for example social valuations of environmental quality or valuations of life and health saving activities via a demand revealing process in a manner that will improve the efficacy of American representative democracy. Several papers at Annex B provide the basis for better understanding the objectives underlying the proposal and for providing more accurate and unbiased information through which sub-national governments and the private sector can assist in realizing national goals. A second set of readings addreses some of the fundamental assumptions about institutions that might be raised in conjunction with any serious pursuit of the approaches laid out in this annex. Two articles -- by T. N. Tidemann raise (1.) particular questions about property rights as opposed to the use of liability rules in moving towards effective application of these ideas and (2.) the need to think about the procedure in a broad context of how to most effectively finance the collective provision of local goods , which may rely to a greater degree than at present on land value increments rather than traditional sources of compulsory financing. We emphasize the importance of this work, in order to suggest that many of the criticisms of the efficacy of the demand revealing approach refects a misunderstanding of the nature of urban externalities and the institutions that woulkd better cope with these externalities. Further, the solutions that we offer as more practical means of addressing current problems may appear to have shortcomings because people are not seeing the larger picture. We will not dwell on these problems but merely point to them in contexts of the particular remedies we suggest as practical solutions to a number of ver immediate and practical problems.

In the context of the overview paper, we believe that the approach can be fully understood in the context of the three applications we have chosen to highlight. We do this in the context of a regional development setting contractual, incentive-compatible approach to airport privatization. Here we define more precisely the contractual, incentive-compatible approach, using one of the incentive-compatible rules called demand revelation, elaaborating on how the demand revealing approach drives social decision-making and the private market towards socially desired ends in specific airport privatization contexts. We will use the expository technique of a parable, similar to those contained in Clarke (1977, 1980), to demonstrate how the political demand revealing governance approach (Clarke, 1977 and others) becomes usefully married to a market system that will achieve socially desired results. The results, illustrated in the context of airport privatization, are generalizable to many other privatization and contractual governance settings._/

1.1 Airport Privatization and Governance.

Most of the myriad of problems described in "Advancing the Debate" result from the absence of incentive-compatible contractual structures and decision rules that can get the parties from here to there in a sequential adaptive way, particularly in the face of a complex web of intergovernmental fiscal entanglements and policy/regulatory conflicts over development (including air transportation) and environmental goals. It is also difficult to shape proposals for airport capacity expansion, financing, and pricing (e. g. user charges on the travelling public and airlines) in a mutually agreeable way and to regulate the system from a standpoint that serves both economic and related social objectives (such as the safety and security of the travelling public) in a way that is compatible with perceived national air transportation needs. The national system is also highly interconnected (as is the case of the air traffic control system) and any institutional arrangements for governing it must take account of this feature, consistent with goals emphasized here relating to privatization and local control of airports.

The shaping of any important regional or local air transportation policy is usually or often a complex intergovernmental (Federal, State and local) undertaking, one fraught with conflict over distributional goals inherent in our system of interest group pluralism. Based on this perception and previous experience, this paper sets forth an approach to intergovernmental and interest group decision-making that can translate governmental and interest group preferences into efficient results, at least when compared with results that will obtain if we continued to rely solely on current institutional arrangements driven largely by the search for distributional advantage.

We illustrate our method in the context of acheiving airport privatization in a regional settings serving a variety of national economic, environmental and social goals. We start with a hypothetical airport in this national system currently valued at about $50 million (or about $19 billion for the nation's 50 busiest airports). In the context of our hypothetical airport, we construct a possible privatization planning process that could be applied to larger regional airports as well as smaller airports, those which may be less subject to the intergovernmental and interest group conflicts we perceive.

The privatization process we describe is conceptually similar to a divestiture, as as that which was carried out in the early 1980s for AT&T. We believe that the benefits of such a privatization could be very large, perhaps as large in relative terms, as those flowing from the AT&T divestiture -- a two to threefold increase in share values for many of the regional companies as compared with the market value of AT&T in November, 1983 at consummation of the divestiture. In the case of airport privatization, however, we are addressing the more complicated problem of best maximizing the potential financial and social returns (which we can only assess now in qualitative terms) through a process that is efficient, distributionally stable, and individually rational. _/ In simple terms, this means making society a lot better off while making all parties affected by the transaction also better off (always a difficult objective to achieve, but one that can be approximated for broadly defined governing bodies and constituencies).

The demand revealing governance method gives each major party in the governance system --Federal, State and local government, an airport operator (selected in a demand revealing "competition for the field" at contract inception, and potentially other more specialized participants in the governance structure (airport users, local transportation and environmental groups or authorities) a "voice" in the planning process, one that wuii be exercised for the most part, truthfully, and without opportunism and strategic behavior. We say, for the most part, because in some stages of the process, involving some very difficult decisions affecting, say the local environment, the lack of adequate information on relative costs and benefits and the ability to ascribe benefits and costs to particular parties, could result in incentives to strategic behavior in the form of coalitions. We have taken particular pains, however, to design an insitutional mechanism that can minimize this possibility and do, we believe a far better job than current institutions. We also take great pains to design a mechanism that will minimize so-called penalty (incentive) or "Clarke taxes" associated with the demand revealing approach, arguing, as have others, that this is a positive feature of the system in forcing choices that move towards, rather than away from, desired social outcomes. Where, for lack of information and knowledge, such penalties do unavoidably arise, we suggest ways of dealing with them in ways compatible with current governmental financial arrangements, and most importantly, without losing the strongly incentive-compatible features of the demand revealing approach.

1.2 A Demand Revealing Parable: Privatizing an Airport

We describe first, for purposes of ease of exposition, a privatization contract involving three governmental partners (Federal, State and local government and/or a specialized transportation authority like New York's Port Authority). Each party will be initially assigned an initial allocation of demand revealing "voting" points that will, at the end of the process, be translated into shares. The initial allocation could be by mutual agreement or some process of arbitration, described more fully later, by which the initial allocation is measured by past and anticipated governmental contributions or related criteria _/ Here we are simply setting some initial cost allocation for providing a "public good" -- in this case expanded private investment and eventual control of privatized local and regional airports -- and the allocation need only be distributionally fair in perhaps a legal or political sense _/ Footnote (One of the more significant legal issues that blocks current privatization efforts we have studied involves who has paid how much in the past to support the airport in question leading to the question whether proceeds from sale or lease of the airport to private parties will be diverted from further airport improvement to some non-transportation use in the local area -- fear of what is called "the money going downtown". The system proposed here is a method for initially addressing the distributional question and moving toward an efficient resolution of efficient regional air transportation decisions).

The decision or governance mechanism also ensures the provision of critical public good or common user inputs such as the air traffic management and control system, a system that has interconnection features analogous to the national telecommunications or telephone system. Our demand revealing approach provides a framework for making optimum decisions on the configuration of this system, from the standpoint of both its interconnection quality and access for local airport users. While improving access as well as the means by which costs of the National airspace system are recovered and paid for, we seek to improve the quality of Federal regulations affecting airports. Federal regulations today emdody a number of often difficult to reconcile objectives aimed at enhacing the safety of the airways and airports and the security of passengers as well as numerous environmental and social objectives, which at times, may conflict with State and local objectives (Footnote - refer to assurances, airport certification requirements, and such factors as land planning, airport noise, access for the handicapped, labor protection, etc.).

Our procedure at contract inception involves the planning and design of a mechanism for cost allocation and decision that properly recognizes the "jointness" and "publicness" of airport management while encouraging State and local authorities (and private airport managers) to play a greater role in planning the provision and financing of the public goods, including the shaping and implementation of regulatory decisions.

The decision mechanism also addresses difficult issues of "jointness" and "publicness" when the governance structure must determine how to raise and allocate revenues -- whether these be called "passenger ticket taxes" or "user charges". New legislation with strong Executive and Congressional support will we believe be shortly enacted to permit local public airports to levy passenger facility charges (PFCs) ranging between $1 and $3 a ticket per passenger gradually supplanting sources of financing through the Airport and Airways Trust Fund or direct FAA appropriations.

The governance mechanism facilitates such decisions as whether a plan, whichmust be approved by the Secretary of Transportation, for levying PFCs, is satisfactory and will cope as well with future airport location and capacity decisions ( how funds should be divided between expansion of existing as opposed to building new airports).

We will now illustrate how such complex considerations of Federal policy and intergovernmental finance can be accomodated in the contractual arrangements governing a local airport privatization. The three governance partners face the decision of soliciting offers from potential operators af an airport, where there is a good deal of uncertainty about the future shape of Federal policy affecting this and other privatizations Will, for example, the privatized airport (more likely a joint venture between the local government and one of the potential contractors) be given acess to PFC financing, qualify for tax exempt bond financing, and/or receive entitlements from ticket taxes generated by the airport? If airport capacity is increased, to what extent will critical air control inputs accompany the capqcity expansion nd how will the regulatory interface relating to a wide range of FAA plan and approval requirements actuaslly work?

Within our adaptive contractual model, we seek a framework in which private bidders may seek as much contractual specificity as possible, leaving to the future better definition of specific rights and responsibilities. However, as these rights and responsibilities become better defined, the initial allocation of points (which eventually become $ shares in the enterprise) also change, so that cost allocations better reflect benefits received The movement from an initial status quo, also involving the acceptable treatment of the values of existing on-site facilities and other scarce inputs, such as slots (landing rights) and gates) to a more efficient allocation, relies importantly on a critical player -- n independent financial Controller -- who attempts to improve on the initial allocation in terms of allocating costs according to benefits received. He may do the job imperfectly, but the process will still result in agreement among the participating parties on efficient allocation arrangements.

We now illustrate the method through its evolution in three stages (1.) an initial contract award (2.) a major second-stage decision on capacity expansion, financing and pricing of the facility and (3,) operating the airport in an environment of mature policy and competitive equilibrium (describing a post-contract stage where the facility in a meaningful sense has become privatized).

1.2.1. Contract stage I: Initial Award

At this stage, we illustrate first how the governance mechanism operates with the assistance of our Controller at contract inception (i. e. initial bidding). We will illustrate with data based on the recent work of others looking at the possibilities of privatization (cite Poole "Privatizing Airports" and Selling Laguardia and Kennedy Airports", Reason Foundation, Policy Studdies # 119 and 208).

Poole's data, for example, suggests we might, at least for expository purposes value an existing airport at a proximate value of about $74.50 per annual enplanned passenger. Based on 1987 actal enplanements, therefore, Laguardia airport might be valued at some amount on the order of $890 million ($64.50 times 12 million 1987 enplanements), Washington-National at $560 million and the Nation's 50 busiest airports at about $19 billion or $370 million for each airport. As Poole points out, such values are often significantly below the current commercial value of the land under the airport if it could be converted to commercial use and the kinds of expansion values we assume do not measure the considerable land use rents in surrounding areas generated by such expansions. _/ Therefore, our illustrations deal with conservastive values where there are also significant departures between financial returns and valuations (even for the facilities and services provided on site) and social valuations and returns.

However, working from such conservative financial values, let's assume an hypothetical local airport valued at about $500 million, generating about $50 million annually in financial revenues and that initially our Federal, State and local partners will have 500 voting points with a value of about $1 million a share. (They have 'paid in" about $100 million in past subsidies, one-half of it Federal and the remainder State and Local). The airport is capacity constrained and warrants a doubling of its capacity and therefore, the partners seek the best airport development and operating proposal that will maximize the social benefit stream as well as possible net social benefits (including off-site benefits and mitigated disamenities such as airport noise).

After some opportunity to study and design their own privatization proposal, let us assume three active bidders each who arrive with an array of proposals and which may have significantly different distributional impacts (i. e. net benefits) to the participants than the allocation under the status quo. Under these circumstances, the Controller is empowered, within limits to reallocate voting points or shares contingent on the selection of a bidder by the three governmental participants. The contracts are also screened, prior to bidding, as to their legality, and consistency with Federal, State and/or local policy , but may have complex options or renewal clauses based on anticipated changes or more precise definitions of law or policy which could alter anticipated outcomes and private/social valuations. Based on the best objective evidence available, the Controller then seeks, within the constraint of reasonable measurement costs, to determine what reallocation of points would be most likely to result in unanimous agreement upon the preferred contract. A simple illustration is provided by Figure 1 where the initial allocation at point A reflects an equal division (Federal vs. State/local but a particular project proposal to double investment in the airport results in a division between points A and B such that, at the margin, benefits will be equal to costs. As the standard demand revealing literature shows, this division may be imperfect but if it is, the parties will still arrive at the efficient result (perhaps point C through a demand revealing procedure which generates a penalty or incentive tax based on the expressed truthful preference of the parties. To achieve this in result in the application here, the Controller is trying to approximate the same result by reallocating the initial distribution of voting points and his incentive to do so, elaborated later, will reflect gains from the decision, administrative and decision costs, and any penalty taxes.

The Controller could have also proposed any other distributional that would have involved no measurement or decision costs whatever. He could have suggested that parties be compensated for past subsidies ($100 million) and that the remaining present anticipated value of the existing facility (rectangle AXYZ) be subject to a throw of the dice. Either rational measurement or such a lottery will do a better job of overcoming actual and potential conflict over the "initial division of spoils" so that the parties will focus resources not on capturing these spoils but acheiving efficiency, moving to point B or even point C. Current airport privatization debates, however, focus a good deal of attention on the division of the rectangle AXYZ because of its size relative to private and social gains to the right of this rectangle that we will describe shortly. In short "disributional rent-seeking" dominates the search for efficiency. (Cite Tullock and rent seeking literature). _/ As suuggested by Tullock and others, one of the chief advantages of the demand revealing approach in the way it mitigates incentives to rent-seeking and drives the search for efficiency, a point we elaborate at some length in this paper.

Let us now deal with the capture of social and private returns in the similarly sized rectangle at the right, defined by ABDZ. In our our hypothetical airport, let us assume that the market in the form of the several (three) bidders anticipates a potential doubling of the current value of $500 million (to $1 billion in present value) with a present value of capacity expansion and other input costs (including profits and any perceived private monopoly rent) of about $250 million (1/2 of the rectangle ABDZ). The bidders, of course, have an incentive to bid away any excess returns (the remaining half of ABDZ) that can be captured off-site and also maximize social valuations (for off-site impacts as measured by the triangle DEZ) at least from the perspective of the governmental parties making the contract selection. At this stage all these costs and returns are merely anticipated, and the alternative contracts are structured so as to best ensure that the gains will be realizable by the governmental parties (i. e. the governmental customers in our transaction). Customers, however, may have no concrete idea of their preferences, and it is the job of our contractual process to help better define these preferences over the course of the contract, leaving the customerts the option to return to point A and incur some costs in doing so. To the extent that the inclusion of parties will improve social valuations , there is incentive to broaden participation balanced by the costs of doing so. At this point no costs (or costly asset exchanges or complex valuations of existing assets) are being incurred other than the time and effort of potential winners of the contract to define a contractual process that will best appeal to the governmental participants. Therefore, we assume that the winner of the initial contract award will simply enter an initial management contract at some competitive level (say to run the airport at 2% of gross receipts which are assumed to be 10% of existing value or $1 million in annual management fees). For expository purposes, let us also assume none of the bidders is seeking to develop off-site land (as is usually the case with such contracts), and that after contract and completion of several steps in the investment process, the winning contract will take a competitively determined number of shares (say 250 shares or voting points measured by the $250 million in anticipated recoverable competitively determined returns, leaving a potential set-aside of another $250 million (or 250 points) for the government partners or other government units (a regional transportation authority or specialized constituencies that could guide decisions on the deployment of these resources in the most socially desired ways. (One might initially question whether any private participant would enter a contract with 250 voting points relative to a total of 1000 being available to government or specialized constituencies. As we will demonstrate, the number of and allocation of votes, and who may have the majority vote is of of general irrelevance in the demand revealing process, unlike the predominance it has had in the creation and governance of existing government/private enterprise in the U. S. -- cite Musoff, "Uncle Sam's Private, Profitseeking Corporations").

At this initial selection stage, we do not envision, however, participants actually invoking the demand revealing process, but rather agreeing on selection of ther best contractor. However, to ensure this optimistic result, the threat of demand revealing must be available and credible. The Federal representativ, for example, must be viewed as capable of delivering a point preference even if he or she would prefer to avoid the intermural squbbling within the Executive or with Congressional representatives in trying to determine the shape of the preference. And by the nature of our contract which requires better definition, any shaping of Federal preferences will become better defined as potential contract specifications are translated into concrete reality at the later satges of contract. Yet the potential for a demand revealing vote at this stage is critically important in motivating the Controller to avoid the kind of distributional competition over division of the spoils we described with reference to rectangle BDEF. We describe the possibility with reference then to a hypothetical vote and how this potential threat motivates a process of efficient cost avoidance, including distributional competition, introduced by the market (i. e. the three potential bidders). To avoid distributional competition, we must permit our Controller to reallocate voting points or shares as these relate to specific contract specifications. Table 1a assumes actual outcomes in terms of preferences for each of three contracts (where differences in expressed preferences for one contract relative to another at this stage of contract are likely to be relatively small in relation to the anticipated $1 billion value of an airport or the $250 million incremental social value from doubling the airport capacity). In the example, the Federal and state authorities have voted a total of 20 points in favor of the winning contract. The state would have preferred the second contract (which is next highest in value) by a difference of 5 points which is different by the same amount as that favored by the locality of 5 points. The "pivotal" voter, in this case the Federal government, wins by a slight 5 point advantage measured in this case by the preference of the State whose choice would have dominated in the absence of the Federal vote_/. As a result, 5 points (the difference between 10 point votes for contract 2 and 5 points for contract 1

could be levied as a penalty against the Federal governement and could be removed from the control of the participants (a delicate matter to be discussed later).

The importance of the Controller's function is again a critical one -- to define the rectangle ABDZ so that the initially anticipated social gains ZED can be realized (otherwwise, as is usually the case, they cannot begin to be realized). Given the initial state of play, not a lot of effort will go into getting to C, where the relevant triangle EOP reflecting true social gain from the socially optimum scale of investment is even smaller. However, as we will argue, the availability of demand revealing is again critically important for the move from A to B and close to C at further contract stages, avoiding costly distributional conflicts in any attempt to move beyond A or to set such a stage in conflicts over the initial distribution of AXYZ.

In moving to define a process for capturing the social gain in the triangles as well as those to be gained on-site, it would appear that governmental preferences would be largely shaped by factors other than strict maximization of social gains in the sense of objectives that might ideally guide a decision on a public investment project. Preferences, at the Federal level at this stage might instead focus of how it injects private capital and the provision of scarce capacity inputs and personnel in a Federal budget-constrained world, compensates for past subsidies, or provides access for future tax subsidies and future trust fund entitlements. In other words, there may be an interaction of economic efficiency and budgetary objectives as well as particularized aviation goals, including distributional goals that shape Federal and other governmental policy. In turn, competing proposals may have different implications and distributional impacts in terms of who pays for what that may shape the preference structures of the other participants, as well as a variety of more subjective preferences or bureaucratic/political process goals that would not be captured in any objective cost-benefit calculus. The objective, however, is not to measure the preference structure but rather to avoid distributional competition and to encourage a process of more specific preference shaping by parties more closely affected by airport decisions Therefore, at the initial stage, the Controller simply trys to reallocate points in response to perceived agendas that introduce distributional competition in ways that will acheive unanimity or zero penalty taxes. Such a result is illustrated at the bottom of Table 1a. where the the Controller reallocates 5 points between the State and federal government, based on a perceived distributional advantage favoring the State relative to the Federal government when contract 2 is compared with contract 1. We shall illustrate later (section 1.3) the compensatory features of demand revelation -- when the point reallocation should provide explicit compensation and when it need not.

Absent the possibility of winning distributional advantage, we expect contractual offers to be driven competitively towards preference shaping in more constructive ways. This may also have an important process dimension that helps to shape governmental preferences not toward simple short-run or narrowly focused budgetary or specific goal related concerns but toward encouraging processes that will, in fact, make effective tradeoffs among these specific goals. With this somewhat heroic assumption, we will assume that we have initially negotiated the world of somewhat diffuse and uncertain intergovernmental preferences and the barriers that it initially presents to a world of private contract governance, and enter a world where the airport must take concrete shape, be financed and prices must be fixed in a socially desired way. A second-stage contract must be specified that comes to grips with the maximization of off-site gains both to the locality and the travelling public (the triangle ZED) so as to maximize competitively determined social outcomes through efficient second-stage contractual arrangements.

1.2.2. Contract stage 2: A Decision on Airport Capacity, Financing and Pricing.

The initial contract award described above reflects, of course, a good deal of uncertainty over critical elements of implementation with respect particularly to airport capacity, financing and pricing. We consider two critical aspects of implementation that will reduce the uncertainty (1.0 a process of contracting with users and the governmental participants to arrive at decisions and contractual agreement on pricing and cost recovery (revenues to be recovered on site and (2) a contract regarding off-site effects with local specialized constituencies, represented here largely by the regional authority

1.2.2.1. Pricing and Cost Recovery

The winning contract at the initial award stage sets the parameters that will shape, in turn, at this second stage, a preferred contract and what returns would accrue to the contractor with that outcome as well as the general obligations that will flow from an investment program to double airport capacity with an infusion of about $250 million in capital, related inputs (including subsidies) and private competitively determined returns. We now want to demonstrate how our process competes away subsidies and artificial tax advantages and defines final prices for air transportation in relation to user constituencies (the air transportation industry). We then turn to regional and local issues, those involving how the process deals with off-site externalities, positive and negative. In Figure 1, we have described about $250 million in social gains to be captured on-site. We will rely on an actual Federally chartered air transportation group that Congress and the DOT have put in place to guide decision-making on the employment of PFCs. Our winning contractor has outlined a broadly conceived plan to involve this and locally constituted user groups (particularly the general aviation community) to help develop an expansion proposal and airport financing/pricing plan that will meet Federal standards (particularly those prohibiting "unjust and descriminatory charges" that is also as efficient as possible, and that will avoid wasting resources in conflicts over the distribution (the winning contractor wants results that are efficient, distributionally stable and individually rational). He also has some property rights allocation problems at the particular airport in the form of slots (the actual rights being somewhat ill-defined but having some apparent significant market value at several hundred thosand dollars each as well as gate preferences for certain airlines that are written in to contracts with the existing governmental airport operator. He wants to win early second-stage approval of his contract by overcoming these entitlement issues, including a cost/recovery pricing package that will satisfy DOT. The Federal advisory body is sympathetic and helps our winning contractor to further refine an imaginative element in the winning award -- the national body of user groups will get entitlement to a share of the anticipated $250 million (or 250 voting points) in return for agreement on slots and gates that will be negotiated among themselves giving then the contractor the right to allocate them, subject to disagreements that can be initially addressed by the Financial Controller in reallocating points (in this case between the contractor and the user group) and subject, if needed, to a demand revealing vote.

As we maintain in "Advancing the Debate", we see no economic reason why a bargaining result of this type could not specify a competitively determined result, if first, care is taken in defining what exists at point A in terms of existing prices and cost recovery (which our Controller would help define) and perhaps subject any remaining costly disputes over the distribution to something easily administered like a lottery, and then define cost recovery for moving from A to B in terms of a mix of charges, including PFCs that will obtain the most efficient results and the highest financial gain for the airport. Our winning contractor has an economist and was told he could actually get peak-load pricing that absolves it of the fear that peak load would be everywhere at all times of the day. Consistent, however, with permissive guidelines on pricing issued by the DOT, the winning contractor discovered a pricing algorithim that constrians the definition of a peak-load price to a careful definition of "marginal delay costs''(expected and actual) which appears administratively feasible and incorporates the demand revealing penalty approach when expected values depart from the actual values of the airline users. (Cite R. Dolan, "Incentive Mechanisms for Priority Queueing Problems", Bell Journal of Economics, 1978).

After significant intramural squbbling within the Federally charted group and also with local general aviation, there is agreement on the general parameters of a pricing/cost recovery scheme although any final agreement is dependent on Federal agreement on the suggestion that a portion of the air transportation share should be dedicated to a national air transportation privatization fund to acheive a similar result at at least 50 of the largest airports or be dedicated to infrastructure and service improvements as part of the trust fund. The division between the national air transportation uses vs. local uses must also be resolved in allocation of the 250 point "set aside" that has been competitively determined at contract inception. Accomodation of regional general aviation requirements must also be made and resloved in the local package. The financial plan, however, provides for full cost recovery of some $500 million on-site without complicated contingency clauses for reopening pricing issues in light of economic circumstances that do not conform to contractual anticipations in specific clauses of the contract involving inflation, interest rates, etc. Economic advisors to DOT have blessed the contract, however, as a model that might be replicated elsewhere, suggesting that in the specific regional context, there is no perceptible element of local airport natural monopoly and no need to specify more detailed contracts for better ensuring the realization of efficient marginal cost pricing in such a natural monopoly context (cite Clarke, 1980, chapter 5 on acheiving such results with an economic natural monopoly).

In such a context, the process will define a set of cost allocations for the facility at its optimum, market-determined capacity and price that capacity so as to achieve the highest competitively determined financial returns. The user groups will help achieve this result for a share of these returns unless they can generate alternatives to change the distribution to obtain more from the taxpayer or through government subsidy. If, however, an alternative user plan was advanced as superior to the contractor's plan, the controller would merely adjust the user's points to compensate. It could also adjust Federal points if one plan included tax subsidies and another (the contractor's) did not. As in the case of our earlier illustration, the point voting or share adjustment remove the incentive to seek distributional advantage in pricing and cost recovery.

1.2.2.2. Local Specialized Constituencies and Off-site Effects

In a simple environment, we might have reached the third stage of a mature stable competitive equilibruim, that could satisfacorily govern a smaller airport in which case we could describe the final consummmation of the privatization. However, in a larger regional setting, we must deal with the competitive pressures of constituencies concerned about land use effects, noise, compatible (mass) transportation, and capture of positive land rents from airport development. To simplify our exposition, we will simply assume that the more specialized interests are represented by a regional transportation authority who will get the remaaining share of financial returns generated on-site (the remaining portion of the $250 million and 250 voting points or shares). Agreement must, of course, be reached among the three governmental parties and the regional authority and the user association on the final allocation of the set aside and this is ideally accomplished by having the Controller ensure that costs (including all recoverable rents now measured by the anticipated $500 million in rectangle ABDZ) are allocated in relation to benefits received which involves determining how the final contract benefits national aviation relative to local interests. This will best ensure that any invocation of the demand revealing rule will minimize the risks of penalty taxes and that the five parties in the transaction will acheive all the gains from the capacity expansion. However, as before, the Controller must avoid stategic attempts by the parties to compete for the distribution, including efforts to delay decisions in hopes of acheiving a distributional advantage. Nevertheless, delay could also have socially productive results in that, for example, a better alternative plan more comptible with the airport capacity expansion or one supplanting it may be in the local interest. At the initial contract stage, the participants may have decided also that some weight should be given preferences for the status quo.

These complications are illustrated by departures from a ideal result when the Controller has failed to achieve the most desirable division or when possibly preferred alternatives have not been put on the contractual agenda. Consider the preference of the Regional Authority to defer a decision on the capacity expansion now with a 3 point vote for the status quo (for a year) as in Table ___. The other participants perceive the loss of $50 million in gains (10% of one year's benefit by waiting, so a local urge to delay would not prevail, nor would it cause generation of any penalty tax. However, one might have wished more attention to the potential dissatisfaction by the regional authority and possibly penalized the Controller for the 3 point vote that did not prevail.

Also, if the dissatisfaction about moving ahead now had been considerably larger, say 30 point votes rather than 3, in the absence of better decisions on land use planning and acceptable noise mitigation planning associated with a capacity expansion, more decisional resources might desirably be devoted to coping with these issues. Alternatively, the Controller may perceive some attempt to obtain distributional advantaage such that he reallocates perhaps a score (20) points away from the regional authority to the other parties. As shown at the bottom of Table 2, this reallocation results in no votes changing the outcome. Thus there is no reduction of the Controller's bonus as a result of penalty taxes being generated, although votes in favor of the status quo remain. However, the latter could be included in an overall performance system to motivate the Controller to seek the objective of efficiency. distributional efficiency and unanimity. The performance system would set a bonus which could be some fraction (say 1%) of:

-- gains to the participants as well as accounting costs he is responsible for measuring (the entire rectangle XYDB0 as well as all rents captured on-site) with emphsis on the incremental gains within the rectangle so as to encourage consideration of maximizing also the triangle ZED.

- minus votes in favor of the status quo (10 points in the above example)

-- minus penalty taxes (reduced to zero in the above example as the result of some costs, including poential costs of challenge, in making a point reallocation

-- and minus administrative costs

Such a method is conceptually similar to giving weights in cost-benefit analysis to distributional considerations and one, of several possibilities, for guiding efforts by the Controller to try to reach as much unanimity as is desired by the participants, given the costs.

1. 2. 3. The "Privatized" Airport at Mature Equilibriumu

We have demonstrated how two very critical decisions to facilitate a privartization could be arrived at through a demand revealing approach. To maximize the social value of the contract, however, many other operational decisions could be subjected to demand revealing pressures motivating the contractor to find the lowest cost way of acheiving social objectives. We illustrate with respect to supplying critical inputs where there are joint responsibilities with the Federal government and the implementation of Federal regulations where there may be the availability of cost-effective alternatives. We discuss to governmental regulation in this context, followed by airport policy when our privatized airport system becomes fully "securitized" in the financial markets.

1.2.3.1. Federal regulation and Airport Privatization

As outlined in "Advancing the National Debate", this approach was introduced in the original OMB reports on government information management decisions as way of ensuring that more effective investment decisions are made when joint gains can be made between the government and a private carrier. The method is easily adapted to our airport.

Take, for example, the case where the airport entity determines that more tower support facilities (personnel and/or enahancements to the modernized air traffic control system are required at the site in question (but funding is not available). In this contingency, the second-stage contract could permit these resources to be supplied outside the cost recovery provisions of that contract, subject to a controller's determination of how the points of each party should be adjusted, not only the Federal government but air transportation users. Disagreements could be subjected to a demand revealing vote so as to ensure an efficient decision.

The same principle applies to a possible Federal decision to command the provision of certain services at the airport facility, say a security enhancement for which resources are not readily available, but warrent no special passenger charges outside the scope of the pricing and financing agreement. The government could still command their provision, subject to the two-step process of adjusting the Federal government's point share with the possibility of a demand revealing vote.

In particular, the decision-making system could be used to seek more flexibility in the implementation of a myriad of Federal regulations and certification requirements aimed at traffic management, safety, access for the disadvantaged groups and other objectives. The contractor, or airline users can advance alternatives that may be preferable to an existing status quo. If differences in benefit perceptions between the Federal government's preferred approach and the alternative are perceived by the Controller, he can adjust points and permit each party to vote. The socially preferred alternative will prevail as long as the Federal government is accurately reflecting preferences not captured in the perceptions of the airlines and the contractor.

Finally, unanticipated new regulations may emerge that may not warrant contract revision. Assume, for example, that Federal preemtory noise standards were enacted, in part as a response to the fear that many localities might enact more stringent noise requirements. Implementing regulations have been difficult to design consisitent with the general desire of the local area surrounding our hypothetical airport to leave the process determined in the second stage of contract undisturbed, even though some in the community do prefer stringent implementation of the Federal preemptive standard. The demand revealing approach to noise regulation is developed here in the context of a potential solution to a real life policy problem at the beginning of fiscal year 1991 (cite the similarity to the OSHA noise standard example in Clarke, "Privatization, Regulation, and the Demand Revealing Governance of Enterprise" (unpub., 1990)

1.2.3.1.1. Case Study: DOT?FAA Aviation Noise Policy: A Demand Revealing Solution.

DOT has formulated a major aviation regulatory noise proposal. The proposal would phase-out all Stage 2 aircraft in the air carrier fleet (about 2000 in number by the year 2003 or perhaps as early as the year 2000). The phase-out would be tied to several related elements to extract a quid pro quo from States and localities to limit the stringency of noise restrictions beyond the capabilities of Stage 3 aircraft. The proposed phase-out might, by some DOT estimates cost the industry upwards of $1 billion, but the cost would be mitigated by a system of trading operating rights among carriers by a system analogous to the existing buy-sell aftermarket at O'Hare, National and in the two New York City airports. Initially, the major issues of interest are as follows:

a. Application of a uniform technology-based standard for ambient aircraft noise. Should we voluntarily repeat or do we have to repeat the mistakes of the 1970s and 1980s?

The imposition of a uniform national Stage 3 standard (applied to a national system of some 1000 or more airports) raises questions as to the likely net social costs of a uniform technology-based standard applied to the industry when the airport site-specific benefits and costs of noise abatement (New York or LA vs. Fargo, North Dakota or Guam) may vary so greatly. Is noise control, during the 1990s, now going to be comprehensively, if blindly, folded into the same philosophy of "nationalizing air quality" a la the Clean Air Act, the implementation of which began during the 1970s and which resulted in the imposition of social costs now being estimated in the hundreds of billions? Is the same philosophy to be blindly applied to ambient airport noise through a comprehensive DOT program (even when noise regulation in EPA has finally been reduced to 1/2th of an FTE), despite the Schulzian annoyance curves and the liklihood that unnecessary social costs might be on the order of only hundreds of millions or one or a few billion rather than the hundreds of billions of economic costs for uniform air quality and true disgreement about how much of that was necessary or desirable. Beyond such philosophical, if contentious questions and the further extension of "the imperial bureaucracy's (FAA's) control into such a major regulatory area are the more operative questions such as how DOT could formulate a reasonable policy for the phase-out that does not prohibit communities with no significant noise problems to permit access of whatever aircraft they like as opposed to the imposition of national environmentalist regulatory philosophies in the same blind manner as before?

If we decide to pursue this issue, we should put together a set of carefully articulated questions designed to get DOT to think carefully about the several dimensions of what they may propose to do. Can the severe noise mitigation problems of greatest concern at major hubs (perhaps the 50 or so busiest airports) be separated from the possibility of continuing to relay on localized solutions at hundreds of other airports? Can we not continue to rely on competition by allowing the latter to utilize older aircraft as they are being phased-out for use in the busiest airports for such purposes as charter and supplemental service, air freight etc, and to permit flights by less than stage 3 aircraft ( such as international flights to a from the U. S. by small, low-cost carriers) into such airports as well? To what extent does the proposal violate competitive norms, not only in the affected industry but legitimate competition among communities and the efficient location of economic activity? Many of these and other questions, ranging from the philosophical to the very operational can be put together so as to stimulate the better shaping of a proposal that could result in achieving the purpose of a 13 year general phase-out without incurring the significant social opportunity costs that are likely to be incurred and that potentially range far beyond the $1 billion that DOT is talking about vis a via the airline (or aircraft manufacturing) industry. In our view, DOT is engaging in a myopic form of utilitarianism blended with what they believe to be practical political reality (designed to trade-off a billion or so of costs to the industry to forestall what is perceived as costs in the several of billions that will result in faster application of potentially greater than stage 3 standards at the busiest airports).

Having raised these issues, a not unnatural next response is to try to be constructive and begin to take the broad objective of the DOT proposal (proposal 1 on phase-out), try to limit its unnecessary uniformity in mimicing Clean Air Act philosohies, and to shape proposals in the second portion of the policy package (proposals 2-4) that would come more directly to grips with the issue of how to best balance (1.) national commerce and air transportation goals against (2.) the desire in many populous localities for increasingly stringent noise limitations.

b.) Balancing National Commerce and Local Preferences: A Incentives Approach to National Aviation Noise Policy.

On a somewhat more constructive note, one can see the possibility for working with DOT to address the overall problem of a phase-out consistent with the diversity of local preferences, not only in (1.) bringing more common sense and flexibility to to the question of hundreds of airports that do not appear significantly noise-constrained but also to (2.) the design of a process for effectively balancing the costs to interstate commerce of local noise restrictions with the localized benefits of noise mitigation. DOT is trying to shape a process (in proposals 2 - 4) that is likely to stimulate significant political controversy, notwithstanding an interesting blend of "carrots and sticks" being considered in the policy package. One could construct a relatively simple demand revealing alternative that would deal with the balancing of interstate commerce and local preferences for noise mitigation in perhaps a more satisfactory, appealing and incentive-compatible way.

Building on the initial effort to introduce demand revealing incentives for efficient information resources management (OIRA, 1983) and related aspects of intra-Federal government decision-making, one could introduce the method into intergovernmental regulatory decision-making with particular application to FAA regulation and the potentially broader evolution towards a regulatory budget and the larger aviation/transportation privatization policy outlined in this paper.

The noise policy initiative offers a major opportunity to reintroduce the proposal to DOT in a way that may help them resolve what could become a potentially mammoth political problem if they were actually try to enforce the quid pro quo they alledgely seek obtain with the phase-out. The basic idea is to substitute for proposal 4 in their policy package (the vaugely articulated "carrot" that permits localities to dip into the new PFCs and other trust fund dollars) another proposal which would help mitigate or even supplant the most undesirable features of proposals 2 and 3 (which involve a heavy-handed DOT or FAA challenging and/or overruling States in the courts and/or administratively when thet choose to move towards more stringent than stage 3 standards and or accelerate their moves towards full-stage 3 implementation. Briefly, the proposal is as follows:

-- Define a status quo which will be the implementation of Stage 3 by the year 2003 (if further analysis indicates this is the most reasonable goal or further point of departure for local option).

-- Specify a demand revealing process for having States and localities weigh the costs to interstate commerce (measured not only by DOT and FAA but finally decided by an independent arbiter) against the benefits to the concerned State and affected localities. The atached Figure illustrates the desired balancing so as to achieve (where a more stringent standard than Stage 3 is desirable) equality, at the margin, between social benefits and costs. Ideally, this can be negotiated with a transfer for the move from A to B of an amount (t) shown in the rectangle. In this case, the State's share of its fixed trust fund entitlement and or local airport entitlements would be reduced by the amount of t which is equal to the cost to the national economy of the move from A to B. If the initial estimates of cost and benefits of moving from A to B were perfectly estimated by our independent arbiter (a sort of independent financial controller/cost-benefit arbitrator), the desired result is achieved unanimously, even though there may be significant dispute over the measures he uses to determine the transfer. If he is wrong, and the optimum point is to the left or right of B (in this case to the right, at C) then some additional amount (a penalty tax as shown by the shaded area) against the State/locality will also be generated, one that will motivated the State and affected localities to make an accurate statement of preferences, so as to arrive at an optimal social decision.

A discrete case example (table 3) shows the results when three governmental bodies are trying to arrive at the best decision over three alternative states. In the absence, of a transfer, the preferred social optimum (the status quo) dominates but a penalty tax against the Federal government (which could be against the discretionary portion of the aviation trust ) because a more stringent, but moderate, limitation would have been chosen in the absense of the Federal prference). With a transfer, however, of 5 points (or $5 million) against the Federal portion of the Trust fund to the State's entitlement), no penalty tax would be generated.

In simple terms, the procedure motivates the honest estimate and expression of jurisdictional (State, Federal, etc.) preferences and an attempt to allocate costs prior to a decision in terms of benefit tax principles (for any changes from the previously agreed upon status quo). In the table, the State's choice (which may reflect economic development and environmental considerations) would dominate in the absence of an articulated Federal preference but with the Federal vote, a penalty tax of $5 million (or 5 points) is generated (equal to the net cost of others of changing the outcome from what it would have been in the absence of the Federal vote). At the bottom of the table is the result illustrating the arbiter's determination that, on the basis of cost-benefit evidence, the State should be allocated $5 million in cost (reflecting the national opportunity cost foregone) for any decision in favor of a more moderately stringent option.

The use of voting points illustrates the fact that $ transfers need not be explicitly involved for any one of a number of regulatory or other policy outcomes. In the larger system of regulation and privatization, the points are related more explicitly to the assets of specific airports (for which Federal/state/local contributions may have been allocated on a 50/25/25 basis) rather than trust fund enttilements which are used in this illustration. The broader application is explained in an extract from "Advancing the Debate" which also demonstrates how more refined elements of the decision-making system could be used to advance and harmonize a number of often conflicting air transportation goals. For the purposes of noise regulation, a relatively simple system could be designed or experimented with for those States that would choose a way of making noise control decisions which may be more consisitent with Federalism, or to avoid the significant process costs of justifying decisions and or having them administratively overturned through the kinds of systems presently envisioned by DOT and FAA in proposals 2 - 4. Alternatively, we envision a viable regulatory budget approach, the first really operational form of a regulatory budget that derives measures of cost and benefits from the potentially expressed (truthful) preferences of the affected parties in the regulatory decision. As set forth in "Advancing the Debate", such a system need not supplant the usual processes that would govern choice and negotiated decision-making, or simple local decisions in the large number of cases (the predominant number of airports in the national system). In difficult case, the existence of a formal demand-revealing decision rule will, however, drive decision-making towards, not away from, socially desired ends.

Towards this end, one could anticipate a process of designing with DOT a legislative package that could lead to Congressional authorization of a demand revealing regulatory approach for making noise abatement decisions and solving the aviation noise policy problem in a potentially more satisfactory manner. Desirably, one would have the the in preparing an Administration policy and legislative package so as to do the design and analytical work this issue deserves. There is danger, however, that DOT/FAA may want to push their present policy/legislative package on a very fast track so that the issue of National aviation noise policy can be linked and resolved with all the other issues to be settled next session. The liklihood is enactment of grand aviation (FAA authorization) package (involving passenger facility charges, slots, noise and other contentious issues) now taking shape. In any case, the DOT/FAA noise aviation legislative/regulatory policy package provides a major vehicle for opening the door to an incentives-based approach to regulatory decision-making and permitting a desirable degree of defederalization as well as privatization in the shaping of more efficient approaches to airport governance. In particular the regulatory approach described above could provide a model for all aspects of FAA policy involving allocation of trust fund dollars. Federal/State preferences would be adequately reflected through a joint demand revealing mechanism of the kind illustrated above and this would apply not only to noise, but the imposition of safety and security, airport certification requirements, etc. To achieve defederalization and basic State control, some basic "division of the spoils" with respect to the staus quo in existing airports would simply be determined and airport privatization decisions would basically be turned over to the States, subject to satisfactory demand revealing decision-making of the type illustrated earlier. (where federal preferences focus specifically on such questions as costs to national commerce, benefits to passenger security, etc.) when regulatory enhancements go beyond some minimum level at a particular airport site. As stated in our conclusions, airport noise is a promising candidate for obtaining the authorization for experimenting with this policy approach in ways that would preserve a reasonable Federal regulatory oversight role, but restore to the States the basic resposibility for oversight decisions affecting airport development, in ways that would probably result in making more effective economic decisions in a manner consistent with other environmental goals (such as noise mitigation and the like).

1.2.3.2. Airport Governance in a System of Competitive Financial Markets and Efficient Local Decision-Making

We have taken some pains in the development of this parable to demonstrate how, in a compex privatization setting, we can introduce the participation of a range of affected constituencies directly into the decision-making process. As set forth in "Advancing the Debate", the need for such a decision-making structure is unclear when we are talking about smaaler airports or those which could be fairly easily privatized like locally (County) owned airports like that of Albany, New York. Perhaps in the case of the latter, we could introduce the additional demand revealing step of subjecting a choice of contractors to the kind of decision illustrated in Section 1.2.1. (Initial Award). However, in thinking about larger airports (the nation's 50 busiest) individually or the possibilities of large scale privatization, we believe the system of governance we are talking about has much to commend it.

For this reason, we have illustrated the creation of "set-asides" subject to the control of specialized transportation and user group control which would not only reflect ownership and control of the hypothetical airport in question be could gradually become nationwide in scope while being integrated in the existing system of national (and international) financial markets. Any user or user group or affected constituency could buy shares or become a "point of accountability" by simply paying the cost of legally establishing itself as such an entity in which it enables itself to vote points or shares and be subject to vote-point or share reallocations. Suppose, for example, that handicapped users of aircraft and airport facilities do not believe that they are effectively represented in the set aside initially established for our hypothetical airport and wish to establish a special point of accountability, particularly as the system becomes applicable to other airports. However, the system has not yet been fully securitized in the sense that shares in the airports are being freely bought and sold in the international financial markets. In this interim environment of a less than perfect capital market (which we believe will pervade the decision-making environment we have created for some time to come), The initial transportation user group has established a procedure and a controller funcntion for establishing spinoffs or newly created entities either as sub-groups in the coalition or as separate entities. We assume these decisions are responsive to properly determined decision and administrative costs being allocated within the user group coalitions.

Similarly, we expect the same evolution within the context of interest group pluralism affecting our governmental Regional Transporation Authority. Particular groups of affected landowners, environmental groups, pro-development groups, local general aviation groups and others, should be enabled to from specially constituted sub-groups or separate groups so as to exercise decision-making influence over the local"set aside". They need only pay the direct costs of exercising their desired influence and be willing to have their shares readjusted for decisions that reach the agenda where thet reap an identifiable distributional advantage/cost or measurable social benefit.

As the system develops, the set-asides (as would the shares to general government) would become fully marketable (or securitized) or simply traded directly among the various groups. For example, the local government in our example might assign its shares to the Regional Transportation Authority for a fair market value well before its shares become fully securitized in the financial market, retaining only enough points to vote effectively where it thinks general government interests, which may not be effectiviely represented by the RTA are at stake. Thus, with effective processes for introducing political participation, we see what may appear to be potential rents being captured by some narrowly based intrest group or political/bureaucratic governing body becoming effectively competed away in a rather efficient demand revealing political marketplace which would be compatible with the desirable efficiencyseeking features of the financial marketlpace. For reasons not always easily understood (and separate for our implicit justification of our procedure as a means of guiding social resource allocation for reasons of externaliities, jointness and public goods, but where important natural monopoly features may be absent, we see some plac3 for our procedures in a system of highly competitive airports in a rather competetive international financial system. This role is elaborated in a separate piece entitled "The Demand Revealing Governance of Enterprise", illustrating how the demand revealing approach provides a check against the unsirable features of majoritarian or other similar decision rules within the enterprise. Cases such as those drawn from the classic treatises of Berle and Means ansd others illustrate how the demand realing method avoids the pitfalls of efforts by one shareholder group to push financial agendas that serve a redistributional rather thabn enterprise efficiency purpose, and are not easily dealt with by the legal system. Thus we see the utility of the demand revealing apparatus at mature competitive equiulibrium as a safeguard against unreasonable and inefficient regulation as well as a check against internal enterprise redistribution (which may of course derive from the external drive of corporate raiders seeking not just enterprise efficiency).

The method we describe can fit comfortably, and help drive, our national airport system toward social efficiency within the existing framework of interest group pluralism and international competetive financial markets. It can also do so in the existing system of local public-goods decision-making. The authors may not be in full agreement or have yet fully decisded their own positions on this matter other than to observe that we can see created enhanced effiicencies flowing from the system we describe, with substantial enhancements to decision-making through the effective inclusion of affected constituencies. Each, of us, is somewhat bothered by the "coercive" features of demand revealing as we are by any system of coercive taxation. However, considerable thinking hasd gone into consideration of the demand revealing approach as it compares with all known systmems of collective choice from the stanpoint of such criteria as social efficiency (including decision-making and administrative efficiency), distributional stability and equality. (Cite with permission, T. N. Tideman's forthcoming book, Collective Choice). The efficacy and potential role for demand revealing has also been considered from the standpoint of acheiving effective local public choices in an overall framework of taxation that has no coercive element whatsover, If the normative goals underlying such a system were widely shared, which they may not be, the authors agree that would would have a truly efficient and normatively superior means of social decision-making. Cite Tideman, "Efficient Local Public Goods Without Compulsory Taxes" and briefly suggest that local communities that shared a particular normative underpinning could adopt certain compatible arrangements (such as changing the base of local taxes from property to land alone) That would move in the direction of rather ideal arrangements for the provision of public goods.

Before turning to some of the fundamental objections that may be raised against using the demand revealing approach within existing institutions and in the specific context of airport privatization, let us deal with a final implementation difficulty -- that of the disposition of the penalty taxes. As we said before, if such penalties are generated, they cannot accrue to the participants in the system -- including the Federal government. Therefore, they are less easily dealt with than the two set asides we have discussed where we are dealing with a particular desire, impl;icit in the existing Airport Trust Fund, that the mony dosen't flow downtown or into general government coffers, but be used for specific transporation uses. However, the amounts (which we believe would be small and essentially one of several "decision" costs of the system remains as a real resource that cannot be regarded as already spent in making decisions. If they has been generated as part of one experimental local privatization, we could of course design an approach where the resources are used to ensure that other localities benefit or use the resources to assist in efforts at international privatization. Initially, they may also have a low value, in that the points generated might be bought at some fraction of their eventual worth before they become fully securitized in the financial markets. What value they have can be used in ways that retain the desirable incentive features of the demand revealing approach, however, as long as the resources are used for some purpose outside of the Federal and sub-national budgets and the kinds of local and national transportation uses illustrated in our earlier treatment of set asides. Short of simply wasting these resources, one of the authors simply suggests that the resources could be used in countries which exhibit "worst-case" manifestations of despotism, anarchy and poverty -- perhaps first in revitalizing their State-run enterprises through incentive -compatible goverance approaches _/

In any case, recognizing that some worthty purpose would exist, short of simply "wasting' the resources, it is critical from a resource allocational standpoint, to ensure that the penalties represent real resources that can flow outside the system and to ensure that the participants have effective ways of trying to reach agreement so as to avoid such a result. However, the potential for such penalties in the absence of agreement and assurance that the penalties will be simply recirculated through the system is critical. Towards this end, the initial contract must have a "bottom line" for converting voting point penalties to financial value and either a separate dedicated use outside the system or a plan for simply wasting them. In either case, the final job for our controller, the performance of which is also measured, is the "trusteeship" of the penalty taxes which can be recovered, through securitization or other means in the same manner as the recoveries by the governmental bodies, user groups and the firm.

Agreement on the disposition of penalties also recognizes the implication that early choices to dissolve the agreement or return to the status quo can have real resource implications (i.e. result in a draw against existing assets). So when the parties enter a bargain, they must weigh these positive costs against the expectation of positive benefit. The result is a positive incentive to ensure that the initial contract and the clauses allowing important demand revealing decisions at later stages is a strongly positive sum, rather than a zero sum game. The prospect that the governing parties and a firm would agree, for example, to such a contract that could generate resource losses or expected gains that are less than some alternative arrangement or the status quo is precisely the decisional calculus that we care seeking to promote. Weighing the risk of actual resource losses from an airport they essentially own and control, even though the allocation of value among them is somewhat diffuse and uncertain, gives them the incentive to design the contractual safeguards that would avoid costly redistributional games and to motivate arbiters, like the Controller, to make adjustments that will ensure all affected parties are benefitted to the extent practicable.

A "worst case' scenario is a considerable degree of disagreement after an initial award that would be tantamount to a contract termination (where the imposition of penalties on governmental parties might require some clear statutory or other legal basis). In reality, of course, a waste of effort to private parties in designing proposals and accompanying protections is conceptually the same as the waste generated by the governmental participants if they initiate such a process but it is later terminated. Short of a total termination outcome might be outcomes that generate penalty taxes for, say, deferring decisions or moving towards final ones under conditions where there is considerable disagreement, and there is no effective way the Controller can mitigate the agreement by cost or point reallocations, Thus contractual agreement by the parties to initially agree on a division of voting points has the risks of waste of the financial value of an existing airport were the parties to disagree. This is illustrated in the case of Table 1a where the Federal government might have been penalized 5 points (or $5 million) if after, initial contract award, the process went no further and/or the process were terminated. The penalty, in turn, would have to flow out of the system as a cost, to the participants, of making decisions. This apparent anomoly is one of the rather pervasive criticisms of the procedure that is dealt with in the following section on problems with the decision-making process. We will deal with several of this and other aspects of the system, leaving it to the reader to identify other, perhaps related problems, that should be dealt with in going beyond our preliminary assessment to a more thourough assessment, and to set the stage hopefully with an experiment with this or even better incentive-compatible decsion-making structurtes in the real world of airport governance. Section 3 sets forth some preliminary ideas on how the mechanisms might be tested in broad-based efforts to encourage the process of airport privatization and regulatory reform.

2. Problems with the Contractual, Incentive-Compatible Method: A Preliminary Assessment

As this is a working document, we will outline the several problem areas that have concerned critics of the demand revealing approach and how they are dealt with in the application we propose. We anticipate this will elicit further criticisms and suggestions that will help to refine our proposal, lead to practical means of implementing it in specific regulatory/privatization contexts, and set the stage for more definitive concusions and future directions for application to more general policy problems and problems of privatization and social control.

3. Conclusions and Future Directions

As a preiminary document, we outline our initial suggestions for application to aviation noise policy (section 1.2.3.1.) and one or several, perhaps related, privatization policy initiatives.