Edward H. Clarke, USAID, American Embassy/Haiti,

U. S. Department of State, Washington D. C. 20520

0.Introduction

Demand revelation is in search of a Ruler, whose job it is to measure. Accompanied by efforts to better evaluate the method and its alternatives in terms of acceptable normative criteria, the role definition of the Ruler will help "place the method in an appropriate constitution that will limit and define the range of applicability" (Clarke, 1977, Tideman,1977).

With the Ruler, demand revealing systems can lessen the undesirable effects of existing systems that serve as "regulators of coercion", while strengthening the decision-making efficiency of those that will "generate agreement" in a Buchanan-Tullock framework of "conceptual unanimity" (Tideman,1984)

At a practical level, the Ruler can define the application of decision rules in a manner that will balance expected benefits and decision costs against appropiately weighted redistributional costs (Buchanan-Tullock external costs) and Clarke taxes. The Ruler's compensation would be related to his success in achiving this result. Relative to what would occur its absence, the procedure will almost always improve the efficiency and distributional stability of outcomes. Although the constitution may determine which rule will govern which issues, the Ruler may be given discretion to specify the selection of rules and to modify the outcomes of all rules so as to maximize benefits less decision costs and appropiately weighted external costs and Clarke taxes.

To illustrate these Ruler-guided decision environments which combine traditional rules with demand revealing rules, I have recasted some applications drawn from the world of enterprise management or corporate governance They illustrate the potential application of demand revealing in settings where less than perfectly efficient capital markets and assymetric distributions of voter preferences call for voting methods that better reflect the intensity of preferences and would provide protections against rent seeking redistributions.

The method has particular applicability to external (.e.g. regulatory governance of the enterprise in multiple objective settings.

The plan of this paper is as follows: Section 1 specifies the role of the Ruler (a social arbiter) who seeks to establish decision rules that will maximize efficiency and distributional stability. Section 2 describes incentive procedures to achieve these purposes in the context of both general applications and those pertaining to enterprise governance. The concluding section deals with normative aspects of the procedure, relating to applications.

1. The Ruler in a Demand Revealing System of Governance.

In an idealized setting of demand revealing governance, it is appropriate to define the role of my social arbiter through a judicial accounting function, which will complement that of the executive (CEO) and assembly (of

stockholders) or legislative committee (Board of Directors). When demand revealing opposition to the actions of the executive, assembly or Board arises, the Ruler assumes the judicial function of arranging compensation or transfers in accordance with Lindahl tax-price principles.

For a more general treatment of the role of a "neutral observer" in a demand revealing system of governance, see Clarke(1971,1977,1980), Tideman and Tullock(1976).In the specialized setting of corporate/enterprise governance, the judge or econometrician of earlier writings becomes the controller or Ruler of G. F. Thirlby(1946) 1/ The Ruler will have the independent function of adjusting share values or dividends to reflect the estimated values (or benefits/costs of alternative choices) among participants in the decision. While the need to make such adjustments (e. g. to arrange compensation) may not be required on many issues because it is not efficient or consensually required, significant or major issues of governance (e.g. merger, liquidation the introduction of new partners) may require that the authority to do so be vested in the controller/Ruler.

While the exercise of this judicial function within the realm of corporate governance may appear novel, it seems appropiate to rely on someone (appointed at the chartering or constitutional stage of enterprise formation), to control the agenda of issues resolution, and select the procedures by which they will be resolved. I have also cast the Ruler into a role defined in the more general framework of Buchanan and Tullock (1962) and Wickstrom (1986) for selecting the optimal decision rule, permitting demand revealing appeals from the Ruler's choices. Further, I have suggested the particular applicability of this role as part of a system of enterprise governance pertaining to the frameworks recently advanced by Zusman (1983) and Nitzan and Procaccia (1987).

This system of governance uses demand revealing rules to drive chioce based decision within an enterprise constitution that will promote efficiency (both procedurally and with respect to outcomes) as well as distributional stability. When demand revealing is introduced into this framework, the Ruler is empowered to make ex ante determinations of required compensation, which may be altered ex post (Clarke, 1977, p46), but not using information generated by the demand revealing rule (Clarke,1971).Tideman (1984) notes the difficulty of accomplishing this result perfectly, and this article is largely devoted to dealing with manifestations of this problem in the more specialized setting of corporate governance. As emphasized by Tideman and Tullock(1976) and Clarke (1980), there must also be an incentive for the Ruler to make correct tax-price allocations, certainly free from the influence of bribery and corruption, accompanied by sufficient information about the distributional consequences of particular outcomes.

2. Applications to Problems of Cost-Benefit Analysis and the Governance of Enterprise

Despite its alleged efficacy as "a new and superior principle for making collective choices" (Tideman and Tullock(1976), the superiority of demand revelation has frequently been questioned from the standpoint of the motivation of a "neutral observer" or Ruler to achieve a Lindahl equilibrium The following applications illustrate how appropriately designed incentives will motivate the Ruler to divide the cost of taking an action in relation to the benefits received, so as to achieve both unanimity and a a Lindahl equilibrium. To the extent that the unanimity result is not achieved, the DR procedure can ensure more efficient and distributionally stable outcomes

relative to procedures that would otherwise be used 2/

In the real world, a decision procedure should not be compared with Nirvana, but rather against the results of what would occur in its absence. However,the procedure may sometimes appear to generate results that are not even collectively rational. Experiments(Tideman, 1983), for example, have shown that the Clarke taxes may actually exceed the efficiency gains from use of the procedure, prompting one to question why a group of decision-makers would be motivated to use it in the first place.

A simple example drawn from a recent example of the attempt to apply the procedure to intra-enterprise decisions (computers and communications facilities) within the U. S. Federal establishment (OMB, 1983) illustrate some of these and other often articulated weaknesses of the demand revealing procedure, and the subsequent attempt to correct these weaknesses through improved incentives and greater clarity in the roles and responsibilities of the Ruler.

2.1 Demand Revealing and Cost-Benefit Analysis: An Example

In introducing application to cost-benefit analysis, we will briefly review how a demand revealing system of governance might operate in a world of full information where the participants are certain about future states of the world. After presenting apprropriate incentive arrangements guiding Ruler administration of decision rules, we will then turn to questions of uncertainty over future states of the world.

2.1.1 The Demand Revealing (DR) Decision Rule

The rule gives any paricipant in a decision the right to change an outcome -- from what would happen in the absence of that person's (group's) paricipation to the person's (group's) preferred outcome by paying a Clarke tax equal to the cost to others of changing the outcome. As demonstrated by our simple example (OMB, 1983), the Clarke tax motivates each participant to represent their preferences truthfully and accurately on a decision whether or not to invest in a $10 million project where the costs have been equally divided among five participants. The net reported benefits (in millions) and Clarke taxes are as follows:

Option A (Status Quo) Option B (Invest) C/Taxes

Party l $0m $1.0m $0.5m

Party 2 0 0 0

Party 3 0.5 0 0

Party 4 0 1.0 0.5

Party 5 1.0 0 0

Under the DR decision rule, Option B would be selected -- the net benefits of Option B ($0.5m) are greater than the net benefits of doing nothing (Option A). However, if either Party 1 or Party 4 had abstained from "voting", the remaining members of the group would have preferred Option A. Because their "votes" swung the outcome, Parties l and 4 are subjected to Clarke taxes (of

$0.5m each under the demand revealing process).

In this example, the Clarke taxes ($1.0) million) are also larger than the net benefits received ($0.5 million). These taxes could also become very high in the case of a "zero sum game" or a case of pure redistribution. Although some might view this as a disadvantage of the process, others view it more positively. As Tullock observes, "this high Clarke tax in many transfers is a positive advantage, as it prevents transfers with no social value from being carried out...and in other cases where it doesn't prevent such transfers, it at least reduces them".

The potential incidence of high Clarke taxes underlines the second essential aspect of the demand revealing approach to enterprise governance. This requires that our Ruler be given the function of allocating "shares" and "costs" so as to minimize their incidence as well as perhaps to weight preferences in favor of the status quo. The previous literature has suggested that this person's remuneration would then be inversely related to the amount of any penalties. This would presumably motivate the Ruler to find the appropriate allocation of costs.

For example, if we assume that the $10 million in project costs had been allocated equally ($2 million each) among the five parties, The result is $l million in penalties because these costs were not allocated in proportion to benefits received. With an allocation of $1 million more in project costs to Parties l and 4 and $l million less to Parties 3 and 5, there would have been unanimity in the decision to undertake the project -- which is a close call, since, as noted above, the total net benefits are only $.5 million or only 5% of total project costs.

The problem is that the Ruler may lack sufficient information to reallocate costs so as to achieve unanimity. If the participants have this information, they could be motivated to report it truthfully. Efficiency as well as distributional stability would also be improved by weighting the objective function of the Ruler and relating his compensation to that objective function.

2.1.2. The Ruler's Objective Function

To achieve efficiency and distributional stability, T. N. has suggested that the members of a collective might establish a presumtion in favor of the status quo by appropriately weighting the Ruler's objective function. The Ruler would, for example, subtract from the net benefits of a decision, some percentage of the value of unintended redistributions and Clarke taxes plus administrative/ decision costs. Discounting unintended redistributions is, of course analogous to giving additional weight to intended redistributions in traditional cost - benefit ananysis See Bohm, 1973, page 111).

The case illustrated here would have the members of an enterprise specify the appropriate discount rate (x%) which would be applied to the Ruler's objective function as follows:

Net benefits less x%(votes in favor of the status quo) less x%(Clarke taxes) less administrative costs

The incentive approach would safeguard against unintended redistributions even in circumstances where no Clarke taxes were generated, motitivating those harmed by a prospective redistribution to express the intensity of their

preferences. Applied to the present example, where individuals have full information and the distribution of net benefits is assumed certain, we assume further that the parties have decided on a rate of discount of 20% to apply to unintended redistributions. In turn the adjusted net benefits of the proposal would be $.5m -.2(1.5m) - .2(1.0m) =0. Taking into account any positive administrative/decision costs, the Ruler would be motivated to exclude the issue from the agenda.

If it is placed on the agenda and not approved, the cost is 20% of the Clarke taxes plus administrative costs. "This assymmetry results from postulating an entitlement to the status quo." 3/

2.1.3. Uncertainty and Multi-Play Decision Making

Although one might readily apply such a procedure to single play events in a world of full information, decisions are frequently taken in the context of uncertainty about future states of the world where participants' gains and losses are assumed to balance out (See Clarke,1987).

Taking first uncertainty, suppose that the net benefit differences in the above example relate to both differences in the liklihood of future events and risk preferences regarding payoffs from future outcomes.

Recently, Nitzan and Procaccia (N-P) have developed a framework for evaluation corporate decisions under uncertainty that might help address our dilemna.

The existence of a situation where expected net benefits exceed external costs/Clarke taxes might, for example, motivate a greater degree of search, so that their is better information on which to base the likelihood of events determining expected payoffs. This may also be accomplished, as N-P have suggested, by assigning search functions to committees or an individual with more knowledge or decision-making capacity. This will better ensure that the expected benefits of decisions will exceed external costs, at least in expectation.

A related way to approach the problem (set out in detail in Clarke, 1987) is to approach the problem probabilistically. Once the Ruler has conducted sufficient search (equating the marginal gains with the costs), the issue must be decided. To do so, he can allow a traditional vote, which in this case is indecisive, or simply throw a coin, with the priviso that those on the losing side can be compensated at the end of the accounting period. (The procedure simply establishes a system of compensation for those whose estimated benefits do not effectively balance out over a range of issues). In turn, Clarke taxes generated on other issues are used to compensate them. The participants can always have resort to a DR vote, but they lose the right to be compensated probabilisticaaly at the end of the period. They must also pay the decision costs (including costs to others if their DR vote does not, in fact, change the outcome).

More broadly, these Ruler-guided modifications would be appropriate whenever the assumptions underlying efficient results under some traditional procedure (e.g. majority rule) are violated 4/. As N-P suggest, majority rule requires homogeneous decision-makers, equiprobable events and symmetric priors. In such cases, efficent outcomes would also be determined under demand revelation without generating Clarke taxes. When these conditions are violated, we need a means of motivating the Ruler to seek more infomation and alternative, more

efficient means of decision-making. Ideally, this motivational calculus would include "success indicators" as well as the threat of heavy potential Clarke taxes that are apt to rear their ugly head when the redistributive impacts of a decision are significant. Hopefully, these taxes would motivate the Ruler to not only arrange proper compensation but to search for alternatives that will generate greater positive benefits net of Clarke taxes. These and other aspects of the incentive calculus of the Ruler will be made clear in the following discussions of more "momentous" corporate decisions and the pursuit of multiple objectives.

To demonstrate further the particular application to problems of corporate governance and the role of the motivational calculus in the Ruler's choice among decision rules, let us now take a basic problem of arbitrary redistribution condisidered by Berle and Means (1932) 5/

2.2 Berle and Means: The Case of Standard Gas and Electric

This leading case related to a corporate charter providing for Class A non-voting) and Class B stock. The Class A common was redeemable at $32.50 per share, was entitled to be paid $25.00 on liquidation in preference to Class B; and was entitled to receive a preferred dividend (non-cumulative) at the rate of $1.50 per share. After this dividend, Class B was entitled to receive dividends at the rate of $1.50 per share; and thereafter the investors were permitted to declare dividends at the rate of not more than 50 cents per share per annum to the holders of both classes. If excess income was to be distributed thereafter, it must be distributed so that the holders of Class B shares should receive four times the dividend declared on the Class A

The Standard Gas and Electric Company ,which owned the great majority of the Class B shares, proposed an amendment to the effect that, whenever Class A and Class B stock had each received dividends of $1.50 per share, all dividends should be declared share and share alike between the two classes; and likewise the permission to redeem Class A stock $32.50 per share was eliminated. Obviously, this cut down the participation of Class B stock.

A dissenting holder of Class B filed a bill to restrain the passing of the amendment but the Court allowed it, in part because the overwhelming majority of Class B shares (that is, Standard Gas and Electric. was prepared to accept it) and this would permit additional capital to be obtained by the issuance of Class A stock (which the corporate management insisted was in the best interest of the corporation.

Consider how how Ruler would treat such a decision. If he goes forward with the Standard Gas proposal via a traditional vote (even with a qualified majority), he risks demand revealing appeals on either side that might generate significant Clarke taxes. Alternatively, assume the representive of the disadvantaged group convinces him that a transfer is necessary in order to minimize votes against the status quo and possible Clarke taxes.

Assume further that not knowing the incidence of net benefits exactly, the Ruler makes a reallocation that would require the transfer of #40 million (or $4 per share) to Class B stockholders. However, the dissenting holders, who have 10 million shares, believe they will be disadvantaged relative to the status quo by more than that amount, say $5 share. Standard Gas, on the other hand, believes that overall corporate benefits of the amendment exceed $50 million by a (net) amount of $10 million and in consequence, the result of a simple demand revealing vote without (w/o) and with (w) the compensating $40

million transfer would be as follows:

Status Quo Pass Amendment C/Tax

w/o w w/o w w/o w

Standard Gas 0 0 $60m $20m $50m 10m

Class B (all) $50m $10m 0 0 0 0

Note how the incentive calculus, described earlier, comes into play here. Without the transfer, the adjusted net benefits of the amendment are highly negative (e..g. $10m - $10 m -$10m = -$10m). With the transfer, they are increased to $10m -2m -2m = $6m.

In order to increase net benefits further, the Ruler could also seek alternatives to such an arrangement. For example, he could propose to simply issue more Class A stock with some alternative distribution of dividends that would not significantly lessen the participation of Class A holders as a second alternative to the Standard Gas plan. Assuming that the 10 million shares would be voted by dissenting Board members as follows:

esult of a demand revealing vote would have been as follows:

Alternative

Status Quo 1 2 Penalties

Member 1 $2.0m $ 0 $6.0m 0

Other Class B Holders 8.0m 0 24.0m 0

Standard Gas 0 20m 10.0 m 0

Total $10m $20m $40m

The result is an improved capitalization plan for the enterprise without significant dilution on the other Class B holders. Further, the adjusted net benefits of alternative 2 relative to the status quo are now $28 million. In this case, the Ruler has been motivated to conduct a search process that would avoid Clarke taxes and would as suggested later, also be rewarded for for arranging further voluntarily negotiated transfers that could reduce them further. Moreover, he will seek to find that alternative that will serve organizational efficieny objectives without generating uncompensated external costs requiring transfers. It is important also that he balance the above with a choice of a decision rule that will also minimize decision costs. If, for example, the Ruler could accomplish his objective with a qualified majority rule, he would do so. Unfortunately, with Standard Gas owning the large majority of shares, it is difficult to see how the efficient result could be achieved by any means other than a demand revealing rule.

The Standard Gas case has illustrated the way in which the incentive rule can motivaste efficient search so as to result in optimum transfers. I would also like to blend this with a system, applying paricularly to government imposed requirements, where the Ruler is motivated to assist in agreement on voluntartily negotiated transfers. In all such cases, we want a system that motivates the Ruler and the participants to invest in enough information gathering and search so as to arrive at the allocatively efficient result, taking into account the option of being compensated ex post at the end of an accounting period. As will be shown in the following section, the resort to DR zand the conclusion of voluntary negotiations precludes such ex post

compensation in a way that can lead the Ruler to also determine patterns of efficient regulation and subsidy, motivated again by the appropriate incentive calculus. Of importance also are the procedural costs confronting the participants. It would appear for example, that is disadvantaged holders in the Standard Gas case had not effected a change in the outcome, but this had imposed additional costs and delays on the corporation, the former would bear some responsibility for these additional costs.

2.3 Demand Revealing Governance with Multiple Goals and Objectives.

In addition to demand revealing criteria to guide the choice of decision rules when there is reasonable concruance of goals (e. g. profit maximization), we have the possibility not only of arbitrary redisitributions but also the pursuit of multiple goals and objectives among the participants. Moreover, the interests of labor or the government may intervene, involving the Ruler in a more complex social calculus, interrnal or external to the enterprise. To illustrate further the application of the technique, we turn to the achievement of certain public service obligations (PSOs) through direct government control or regulation of the enterprise.

In this more complex setting, the Ruler's objective function should also reflect the full value of voluntarily negotiated transfers among the participants, designed to maintain a Lindahl equilibrium. To see the basis for this modification of the basic incentive rule, consider the Standard Gas case presented earlier where the Ruler had specified a $40 million transfer that changed adjusted net benfits from - $28 million to a positive $4 million. However, a coalition, wishing to preserve the status quo could have each revealed not $2 million, but rather $5 million such that the net benefits of retaining the status quo were positive (by $10 million() and Clarke taxes are eliminated.

As explained in Clarke (1980), when such incentives are present, there is also the dominating incentive to negotiate additional transfers and this can be reinforced by including the full value of transfers in the Ruler's objective function. Let us consider how this would work in the context of determining regulatory compensation and subsidy.

2.3.1. Demand Revealing Determinations of Regulatory Compensation and Subsidy

The approach set forth here basically extends an idea first advanced by Portney and Sonstelie (1983) who pointed to difficulties posed by efforts to quantify the costs, much less the benefits of regulation. Why not, argued Portney and Sonstelie, have the affected firm identify the true social opportunity costs through demand revealing procedures, which it is motivated to do truthfully and accurately? The government can then determine whether or not to implement the regulation depending on its perception of benefits in relation to these revealed costs.

As Tullock (1977), Tideman(1979) and Clarke (1980) have shown, the efficacy of this approach is enhanced through entitlement to compensation, which can be arranged independent of how rights may be initially assigned. For example, a compensating subsidy may be paid to the firm (because it otherwise has a right to generate an externality such as pollution) or the firm may pay an amount to the government reflecting its efficent buy-out from a government mandated requirement.

The following two sections draw on other work illustrating means of obtaining efficient adjustments to relatively coercive systems of regulatory standard setting in the United States (section 2.3.2.) and for more efficient means of implementing contract arrangements with State enterprise (Section 2.3.3) The illustrations are intended to show how the role of the Ruler (now a judicial/ regulatory agent attached to the enterprise) might be defined, so as to motivate him to make efficent initial determinations of regulatory interventions, using the basic notion of "compensated incentive compatability". As part of this role, and in a manner similar to the Standard Gas case, the Ruler also encourages the affected firm to seek alternative, less costly, but perhaps more cost-beneficial means of achieving compliance (or meeting a PSO) relative to the specific standard or intervention proposed by the government.

3.2 Ruler Determined Entitlements to Compensation: Governing the State Regulatory Apparatus

As an arbiter between the enterprise and the State, the Ruler could essentially replace much judicially determined regulatory enforcement in the United States. The Ruler, whose objective is to obtain an equality , at the margin, between the social benefits (and costs) of regulation, seeks to determine what set of standards enforcement will achieve this objective at the enterprise level. Using a diagram based on Tullock (1977) and adapted by Clarke(1980, p.141), Figure 1 also demonstates departures from the Ruler's initial calculation of this optimum, when the government seeks some higher level of enforcement. When it perceives the benefits would be higher than initially determined by the Ruler, Clarke taxes, in the amount of the diagonally shaded area in Figure 1 would be generated. On the other hand, the firm may perceive its opportunity cost are greater than the ruler has calculated, in which case it seeks a lower level of enforcement, paying the Clarke taxes illustrated by the dotted area in Figure 1.

In the real world of regulatory enforcement, I would envision many Rulers (essentially regulatory enforcemnt authorities, seeking to promote voluntary consent among potentially many firms. Under the existing State regulatory apparatus, subject to the objectives of multiple regulatory authorities, the cumulative opportunity costs of the multiple interventions are difficult to control through traditional means of government oversight. Accordingly, many have perceived the need to fashion something akin to a "regulatory budget" that might somehow mimic the fiscal budget process as a means of disciplining

and controlling regulatory costs.6/

In this context, one can envision a reasonably efficient set of arrangements for ensuring reasonably cost-beneficial regulatory choices, with some constraint on the level of national income consumed by such regulation.

Each year, information would be provided on relative general benefits tied to specific types of regulatory enforcement action, with the level of implicit enforcement standards designed to yield a rough equality between refenues generated from firms seeking relief ( through payment of taxes) and those being subsidized to meet standards that are determined to be highly cost-beneficial, or which exceed legally required levels. To avoid Clarke taxes, the Rulers would seek to equate firm-level opportunity costs with benefits signalled by the regulatory authorities.

Appropriate funds would be set aside (for use of agencies and intervenor groups) who seek modifications of Ruler-determined levels, and any Clarke tax payments would be used to compensate firms subject to multiple regulations, for which the aggregate net negative impact is judged to be the greatest. (see Clarke, 1987)

The system of enforcement of standards would thereby become largely Ruler-determined and since this conveys a rather significant degree of regulatory authority, his determinations would also be appealable in the courts (with the costs of decision appropriately allocated to the winning and losing parties in ways that will not significantly lessen otherwise attractive incentive properties of the system.

2.3.3 Governing State Enterprise

A demand revealing approach to implementing a means of discipling the State regulatory apparatus also extends to the governance of State enterprise. Although this is the subject of companion work,the efficacy of such a contract will be enhanced when we permit the Ruler to provide explicit compensation, and accord he or she sufficient discretion to oversee implementation of the specific terms of the contract.7/

For example, a contract plan for the State railroad in Senegal carries with it certain public service obligations (e. g. maintenance of track, setting up barriers at crossings), where costs may be uncertain, although the division of them between the government and the enterprise may be specified, or when future events (e. g. official tariff restraint) may be quite uncertain or open to negotiations. (Mallon, 1983, p17)

With incomplete contracts, sufficiently precise to permit the Ruler to determine when, and on what terms compensation will be required, we have a means of fulfilling necessary public service obligations through State enterprise in ways that will avoid deleterious effects on efficiency.

In this context, we could also apply the concept of the "regulatory" or State enterprise budget", perhaps in the context not only of single (large) State enterprise, but also to those which operate in several sectors, where the forces of existing or potential competition can also be permitted to work in ways that will improve the performance of the State (or regulated ) sector.

3. Conclusion

Adapting demand revealing systems of governance within the existing State regulatory apparatus and to further encourage the process of privatizing State enterprise will require a good deal of ingenuity. Factor mobility and competition can also play an important role in this process.

Governments in developing countries may be initially resistent to privatization, but they nevertheless face increased restrictions in their access to debt capital, or are interested in exchanging public enterprise debt for equity and in attracting increased flows of direct foreign investment. To make privatization work in a global context, experimentation with new methods of State enterprise governance might incorporate the demand revealing protections advanced in this work. Rational investors, of course, demand such

protections, which would extend well beyond normal "political risk" insurance available, for example, to U. S. investors abroad.

In response, governments seeking to accelerate the process of privatization by attracting new capital might accord new attention to the range of subsidiary protections that could be accorded to joint venture partners, foreign and indiginous, majority and minority alike. In these settings, which ordain the creation of new or revised corporate charters (counting in the hundreds or even thousands over the next decade and beyond), there will exist a potentially ripe climate for the for the establishment of demand revealing constitutions, embodying a range of investor protections that will help stimulate direct private foreign investment. To exploit this potential, governments and investors should invest rather heavily in the appropriate protections. For demand revealing systems, these would define the Ruler's choice of optimal decision rules (including the resort to demand revealing rules) and would clarify the roles and responsibilities of the host government in providing Ruler-defined compensation for the fulfillment of presently anticipated or future public service obligations.

Compensation that is not, in turn, provided might not only reduce the credit standing of the country (a normal practice), but be used to offset the official assistance flows from multilateral and bilateral donars. Similarly, penalty taxes that arise from imperfectly designed compensatory arrangements illustrated in this paper, might flow to those donar agencies -- to be used to further combat the problem of worldwide poverty.

In addition to the attraction of capital, privatization must also deal with the labor market dimension. Companion work, which deals in more depth with demand revealing approaches to labor market adjustment, may have further relevance to the actual "how-to-do-it" of encouraging rational methods of governing SOEs, while privatizing them. More broadly, the difficulty of labor adjustment to economic change has been one of the most significant impediments to privatization of many nationalized industries, both in developed and developing countries.

These illustrative applications are merely suggestive of how the worldwide market for factors -- particularly international capital -- may be used to stimulate interest in incentive compatible mechanisms. However, we have dealt only in passing (and in the context of privatization) with the relationship of the factors of production, as for example, the relation between capital and labor in demand revealing managed institutions. This relationship, and the explicit role of the Ruler, will be addressed in other work concerning, for example, the management of producer cooperatives.

In stimulating more attention to demand revealing institutions in developing countries, we continue to struggle with normative attributes of the procedure. These attributes of the process and other voting rules are being considered in a forthcoming book by Tideman which evaluates rules by the criteria of efficiency (both procedually and with respect to outcomes), equality and distributional stability (Tideman, 1977 and 1984)

In emphasizing the role of the Ruler and the demand revealing efficency of the capitalist or socialist enterprise, this article has also sought to elaborate the means by which demand revealing decision rules, accompanied by appropriately constituted powers invested in the Ruler (including properly trained accountants) can contribute not only to efficiency, but also to

distributional stability.

As for equality, alternative means for improving factor mobility, enhanced by entitlements to equal shares of unimproved land and natural resources (Tideman,1984) provides a means for establishing the essential "fairness" of Ruler guided demand revealing systems of governance This is a direction I also tried to establish in my 1980 book, largely in response to the perceived need for both compensated incentive compatibility and means of addressing problems of fairness and equality through appropriately defined entitlements.

For demand revealing applications to enterprise, "the heart of the matter" continues to lie in somewhat more pedestrian "accounting questions". In this arena, and apart from a larger study of demand revealing approaches to privatization that includes also the governance of producer cooperatives and state regulated natural monopoly (see also Clarke and Tozzi, 1983), companion work elaborates on the use of the method as a means for controlling the current State regulatory apparatus in the United States

More broadly, the method provides a useful tool in the arsenal of those seeking to define the minimal state, while combatting "redistributive churning", the Leviathan or even the plantation State (Buchanan, 1987). In this regard, one might suggest that "charity begins at home" or somewhat more care in the application of the demand revealing approach to State enterprise or to the alleged emerging Plantation States of either the developed or developing world. However, in the often separate worlds of the budget, regulation, and privatization, innovation is moving faster than one might once have thought and who knows when the time for demand revelation might come. With it, we might knowingly evade, perhaps even govern, the Plantation State, rather than blindly let it govern us.

1/ The reference to Thirlby's "Ruler" is not meant to be oblique or to connote undue power in the hands of corporate controllers who would be empowered to make wholesale wealth transfers within the enterprise and between the State and enterprises. Thirlby (1946) defined the Ruler and his relation to the administrators of an industrial unit in terms of an ad hoc authority external to the industrial unit concerned. Thirlby refers to the ad hoc authority as the Ruler "because his function is to measure" (opportunity cost). So too is the function of demand revealing rules and procedures, which are joined here with the Ruler in order to maximize the responsibilities of the enterprise accountant to measure and distribute costs in ways that will maximize business opportunities, profit and economic rent, while reducing inefficient rent-seeking behavior within and affecting the enterprise.

2/Both this and companion work on application of demand revealing decision rules has been driven by an apparent lack of proper specification to the role of the social arbiter in any system of governance. The problem came to my notice as an outgrowth of an October,1983 proposal to the U. S. Federal agencies to use demand revealing decision rules for making certain intra-enterprise dcisions within the Federal establishment. As might be expected, one of the major concerns of the agencies were the distributional consequences of the proposed procedure, irregardless of its efficacy in promoting efficient outcomes. Furthermore, who was to decide the initial allocation of cost shares and might there not be all kinds of mischief if the unnamed arbiter were to fail in his unaccustomed distributional role? While not fully equipped to provide the answers at that time, I have been subsequently motivated to more adequately investigate these questions, and the results are reported generally in this article as well as in the companion work dealing with intragovernmental decision-making and regulatory governance.

3/ I am indebted to T/ N. Tideman for suggesting the incentive rule suggested in this paper (private communication, April 29, 1987)

4/My interest in problems of redistribution in majority voting was originally stimulated by Gordon Tullock (1959) who also pointed to means by which demand revealing technigues might provide safeguards against redistribution and rent-seeking(1985).

5/ Book II of Berle and Means (1932) basically portrayed the loss of shareholder control over management decisions involving; (i) changes in participation accruing to shares, (ii) routing of earnings as between shares of stock, and (iii) alteration of the original contract rights of security holders Legal procedures designed to protect, for example, against unwarrented delutions had become too time-consuming or expensive or, as in the case of the pre-emptive right of shareholders to purchase new stock, experts had determined there was no (satisfactory) way to provide such a right where there were complex classifications of stock in the corporate structure. The Standard Gas and Electric case, discussed in the body of this paper, appears at page 215-216 of Book II.

6/ See Nordhaus and Litan (1983) for an appraisal of the difficulty of the "regulatory budget" concept. The applications illustrated briefly in this article and elaborated in companion work are gleaned from some fifteen years experience working with, or in, organizations that seek to exert pressures

toward efficiency outcomes affecting major enterprises under the direct or indirect control of the State or through its regulatory apparatus. To a more limited extent, this article is also influenced by one year's experience and observation of those encouraging more effective means of oversight and control of State enterprises through changing the role behaviors of financial controllers responsible for public enterprise experience in a French-speaking North African country. All of the country's enterprises had inherited French systems of governance and accounting , which it is altering through significant changes in a National Plan comptable which will also fundamentally change the role of financial controlers responsible for monitering enterprise performance, and reporting to stockholders in all enterprises. In my reflections on these changes, and in conjunction with companion work on the control of the U. S. regulatory appartus, I developed further my notions on implementing a "subjective theory of accounting" and the essential role of Thirlby's "Ruler" in a demand revealing system of governance . I then departed to another French-speaking country in the Caribbean where I hope to carry forward this work.

6/The impetus for much of this work has come from work on problems of privatization. In the North African country where I have been residing, for example, much of the economy is in the hands of some 600 public enterprises which control about half of the production of the industrial sector.

The rationale for State enterprise in this country derives from numerous social and economic objectives, an important one being the desire to attract foreign capital that could be combined with indiginous capital which is otherwise difficult to mobilize in the presence of thin and illiquid capital markets. In spite of this objective, little foreign capital has been attracted, in part from the fears of State control over decisions, irrespective of the composition of ownership and voting control.

In this context, the resort to a demand revealing system of governance could make a good deal of sense in the context of either ensuring efficient outcomes in the presence of multiple objectives or even in pursuing the single objective of enterprise value maximization in the presence of an undeveloped capital market. Decisions on future capitalization could take on many of the characteristics of the Standard Gas case, to read in this context, the government vis a vis joint venture partners. Such a regime would impart both greater efficiency and an enhanced degree of investor protection as well as improved distributional stability in the affairs of the enterprise and commercial relations between Nations.

Berle,A. and Means, G.(1932). The Modern Corporation and Private

Property.New York: Mcmillan

Bohm, Peter. (1973). Social Efficiency. New York: John Wiley

Buchanan, J. M. (1987). From redistributive churning to the

Plantation State, Public Choice 51 (2):241-243

Buchanan, J. M. and Thirlby, G. G. Eds (1981) L. S. E. Essays on

Costs: New York: New York University Press

Buchanan, J. M. and Tullock, G. (1982). The Calculus of Consent.

Ann Arbor: Michigan Univ. Press

Clarke, E. (1971). Multipart pricing of public goods, Public

Choice 11: 17-33

Clarke, E. (1977) Some aspects of the demand revealing process,

Public Choice (special supplemnt) 29 (2): 37-49

Clarke, E. (1979) The search for budget balance in a theory of

demand revelation, in Gordon Tullock, ed, Public Choice in New Orleans, Center for Study of Public Choice, Virginia Polytechnic and State University

Clarke, E. (1980) Demand Revelation and the Provision of Public

Goods. Cambridge: Ballinger Pub Co.

Clarke, E. and Tozzi, J. J. (1983) On information and the

regulation of public utilities, in A. Danielsen and D. Kamershen, Eds., Current Issues in Public Utility Economics,133-147: Lexington: Lexington Books

Clarke, E, (1987) Tideman and Tullock Reconsidered: A review and

comment (unpublished)

Litan, R. and Nordhaus, R. (1983) Reforming Government

Regulation: New Haven: Yale Univ. Press

Mallon, R. D., 1983, Performance contracts with state enterprise

HIID Discussion Paper #143, Harvard Univ.

Nitzan S. and Procaccia, U. (1987) Optimal voting procedures for

profit maximizing firms, Public Choice 51(2):191-208N

Portney, P. and Sonstelie, J. (1983), Truth or consequences:

cost revelation and regulation, Journal of Policy Analysis and Management

Tideman, T. N. ed, Special spring supplement, Public Choice,

29(2); 1-9, 71-77

Tideman, T. N. Property, liability and incentive compatibility:

toward efficient management of urban externalities, Working paper 78-11-2, Center for Study of Public Choice, Virginia Polytechnic and State University, Blacksburg, Va.

Tideman, T. N. (1983) An experiment in the demand revealing

process, Public Choice 41: 387-401

Tideman, T. N. (1984) Toward restructuring of normative

economics, American Behavioral Scientist: 28(2): 233-247

Tideman, T. N. and Tullock, G. (1976), A new and superior

process for making collective choices, Journal of Political Economy, 84: 1145-59

Thirlby, G. F. (1946), The ruler, in J. M. Buchanan and G. F.

Thirlby, Eds, L. S. E. Essays on Cost 165-198, New York: New York Univ. Press

Tullock, G. (1959) Some problems of majority voting, Journal of

Political Economy 67: 571-579

Tullock, G. (1977)Revealing the demand for transfers, in American

Re-evolution, Papers and Proceeedings in R. D. Auster and B. Sears eds, Phoenix: University of Arizona: 107-23

Tullock, G. (1985) Demand revealing, transfers and rent seeking

U. S. Office of Mangement and Budget (1983) Incentives for

efficient information resources management, a Supplemental Report to Managing Information Resources , Second Annual Report to Congress under the Paperwork Reduction Act of 1980

Wickstrom, B. (1986) Optimal Majorities for decisions of varying

importance, Public Choice 48 (3): 273-290

Zusman, P. (1983) Collective choice, pareto optimality and the

organization of cooperatives, Journal Economic Behavior and Organization . September: 185-204