POLICY EVALUATION:

WIN-WIN JUSTICE

Stuart Nagel, MKM-PSO-DSI Center

and University of Illinois

I. DEFINING JUSTICE

Justice is one of the most important concepts in social, political, legal, and general philosophy, and yet it is one of the most subjective. It is easy to define justice with words that are like synonyms, but difficult to define justice in operational terms so one can objectively say that one public policy alternative is more just than another. Looking in a thesaurus or a dictionary, one would find justice defined in terms of fairness, equity, and other words that get at the idea of doing what is right, good, or better in social policy.

One could read whole books on the subject, not just dictionary definitions, and still not come away with any objective criteria as to what justice is unless one is willing to accept the assumptions of the authors of the books. Those assumptions are either on such a high level of generality that they are cliches or synonyms for the concept of justice; or they have built-in biases in a liberal or conservative direction. If so, they are acceptable only if one has a liberal or conservative value orientation to begin with.

Simple examples could be given of principles of social ethics that at first glance do sound like they have objective reality, but at second glance tend to have distinctly liberal or conservative biases. Liberal biases tend to favor the have-nots or the general population in a society. Conservative biases tend to favor those who are relatively well-off.

A good pair of examples are the widely accepted principles of Pareto optimality and utilitarian pragmatism. The basic principle of Pareto optimality is that social policy A is just if it results in at least one person in a society improving in terms of his or her well-being, and all other persons at least remaining constant but not worsening. Social policy A is better than B if two people rather than one person undergo improvements.

The basic related principle of utilitarianism is that social policy A is just if it results in improved happiness for a majority of the people in a society, while allowing the possibility of some people worsening but not below a minimum threshold. Social policy A is better than B if it promotes more happiness for more people.

The two principles conflict partly because one has a conservative bias and the other has a liberal bias. Strict Pareto optimality would consider progressive income taxes combined with government aid for the poor to be unjust because rich people might have a net loss between their costs and benefits. Poor people would have a total net gain that might more than offset the total net loss of the rich. There are more poor than rich, and a dollar given to a poor person may increase more happiness than a dollar taken from a rich person decreases happiness. Such a system of progressive taxes and government aid would be considered just by a utilitarian philosophy given its definition and its liberal biases.

The object of this article is to discuss a different way of conceptualizing justice so that social policies can be considered just or unjust, or can be considered more just or less just regardless of whether one has a liberal or conservative set of values. The object is not to arrive at compromises in accordance with Aristotle’s golden mean or the moderation of Confucius. The object is to be able to simultaneously achieve or more than achieve the goals of conservatives and liberals, including doing well for both the rich and the poor at the same time.

II. THE OBJECTIVITY OF JUSTICE

Justice refers to both procedural rules and substantive rules. Justice is objective if one accepts certain assumptions:

1. The conservative assumption is that the goal to convict the guilty is most important. This partly means to convict about 90 percent of all guilty defendants, guilty arrested, or guilty perpetrators.

2. The liberal assumption is that the goal to acquit the innocent is most important. This partly means to acquit about 90 percent of all innocent defendants or arrested persons.

It is good to exceed both conservative and liberal best expectations simultaneously. This relates >90 percent Convict Guilty. It also relates to >90 percent Acquit Innocent. Exceeding both expectations is good because it would be endorsed by the highest percentage. Assumption 1 is endorsed by 40 percent of the lawyer population. Assumption 2 is endorsed by 50 percent. “Do not know between 1 and 2,” is endorsed by about 10 percent. Exceeding both assumptions is endorsed by about 100 percent, which equals 40 percent + 50 percent + 10 percent.

Just because 100 percent endorse something does not make it objectively fair/good or objectively real/true. The correct phrase is fairer over the previous system, or fairer between two acts, or fairest of a set of acts. For some people, criteria are fair because they are fairly adopted. But if conservatives have a majority, liberals would consider it unfair. If liberals have a majority, then conservatives will consider what they adopted to be unfair. For some, there is no fairness in rules themselves (procedural or substantive), only in how they are adopted.

That which is just is that which super-optimizes:

1. A just policy is one that exceeds the best initial expectations simultaneously of conservatives, liberals, and other major viewpoints on the problem to which the policy relates.

2. An unjust policy is one that does worse than the worst initial expectations simultaneously of conservatives, liberals, and other major viewpoints on the problem to which the policy relates.

3. A policy that is not clearly just or unjust is one that neither does better than the prior best expectations of conservatives, liberals, and so on, or worse than the prior worst expectations of conservatives, liberals, and so on.

III. EXAMPLES

Three SOS or win-win examples can be given, drawing upon criminal, civil, and economic justice. The key dilemma or dispute in criminal justice is based on the desire of conservatives to have all guilty persons convicted versus the desire of liberals to have all innocent persons acquitted. The object is to develop procedures of super-optimum justice where there is improvement in the achievement of both goals over the status quo. One example is videotaping all trials to provide an easily accessible record of the testimony, thereby facilitating greater accuracy in both the convicting of the guilty and the acquitting of the innocent.

TABLE 1. WIN-WIN CRIMINAL JUSTICE

GOALS

ALTERNATIVES

Convict the Guilty

Acquit the Innocent

Easier to Convict the Guilty

–

Easier to Acquit the Innocent

–

Between <C and >L

SOS OR WIN-WIN

>C and >L

NOTES:

1. Examples of rules that make it easier to convict the guilty include 6-person juries, majority votes, limited appeals, and restrictions on calling or cross-examining witnesses. Examples of rules that make it easier to acquit the innocent include 12-person juries, unanimous vote, easier appeals, and less restrictions on calling and cross-examining witnesses.

2. An SOS alternative in this context would be any rule that simultaneously increases the probability of convicting the truly guilty and probability of acquitting the truly innocent. Examples would be better training of jurors and judges on the meaning of various jury instructions or legal concepts and allowing the jurors to ask questions of the judges or lawyers to obtain clarification of legal concepts or evidence matters.

The key dilemma in civil justice is based on the desire of conservatives to protect business firms and insurance companies from liability to consumers, workers, and others versus the desire of liberals to have injured consumers, workers, and others adequately compensated. The object is to develop procedures that will simultaneously save business from liability while saving would-be victims from injury. In the realm of automobile accidents, this may mean public policy directed toward safer cars, safer roads, and safer drivers.

TABLE 2. WIN-WIN CIVIL OR TORT JUSTICE

GOALS

ALTERNATIVES

Avoid Liability & Lawyers

Compensate the Injured

Difficult Liability

–

Easy Liability

–

In-Between

SOS OR WIN-WIN

Reduce Accidents

NOTES:

1. Having a defense of contributory negligence makes it relatively difficult for the plaintiff to establish liability. Having a system of no-fault makes it relatively easy for the plaintiff to establish liability. Comparative negligence is in- between.

2. Reducing auto accidents by improving cars especially refers to having airbags to protect drivers. Improving roads especially refers to having more median strips to prevent head-on collisions. Improving drivers refers especially to more severe penalties for drunk drivers.

The key dilemma in economic justice is based on the desire of conservatives to have individualistic private enterprise versus the desire of liberals to conduct business in a more collectivistic way. One form of super-optimum economic justice in this context is to have title or ultimate responsibility for factories, farms, public education, police protection, and other such societal activities in the hands of government while contracting out the operations for all such activities. Another form of super-optimum economic justice is to have a free market that is made competitive through well-placed government subsidies and tax breaks designed to stimulate multiple competing firms in every industry. Still another form of super-optimum economic justice in the realm of labor-management class-conflict is to provide for relatively high wages to satisfy labor, but relatively low payments to satisfy management. The difference is in the form of a wage-supplement voucher with strings attached designed to provide for the upgrading of skills through on-the-job training and the hiring of unemployed people, especially those outside the labor force.

TABLE 3. WIN-WIN ECONOMIC JUSTICE

GOALS

ALTERNATIVES

Quality Goods

Low Prices

Marketplace Monopoly

–

Government Ownership Monopoly

–

Some of Both or Mixed Economy

SOS OR WIN-WIN

Stimulate Competition

(see Note 2)

NOTES:

1. The marketplace may lead to monopoly or oligopoly by virtue of one or a few strong firms eventually dominating the industry and eliminating or greatly reducing quality and price competition. Government ownership normally means only one firm in an industry or in a region.

2. Competition can be stimulated by public policy through giving seed money to new firms. Also by requiring the sharing (at reasonable fees) of patents, railroad tracks, telephone lines, electric lines, and other facilities. Also from international trade in goods, services, labor, ideas, and factories

IV. PRINCIPLES

These examples lead to three principles of super-optimum or win-win justice. All three principles can be illustrated by all three examples, rather than one example for each principle. The first principle is that an objectively just policy exists if the policy enables conservatives, liberals, and other major ideological viewpoints all to come out ahead of their best initial expectations simultaneously. The second principle is that an objectively unjust policy exists if the policy causes conservatives, liberals, and other major ideological viewpoints all to come out worse than their worst initial expectations simultaneously. The third principle is that a policy is neither objectively just or unjust if it enables one or more major ideological viewpoints to come out worse than its best expectations, although it comes out better than its worst expectations.

TABLE 4. WIN-WIN GENERIC JUSTICE

GOALS

ALTERNATIVES

Conservative

Liberal

Conservative

–

Liberal

–

Neutral

SOS OR WIN-WIN

NOTES:

1. An objectively just policy has been found on the SOS row if the policy scores (1) higher on the liberal totals than the liberal alternative and simultaneously (2) higher on the conservative totals than the conservative alternative. In the above generalized context, that means scoring better than a 14 on both columns.

2. An objectively unjust policy has been found on the SOS row if the policy scores (1) lower on the liberal totals than the conservative alternative and simultaneously (2) lower on the conservative totals than the liberal alternative. In the above generalized context, that means scoring lower than a 10 on both columns.

3. A policy that is neither objectively just nor objectively unjust has been found on the SOS row if the policy scores (1) lower on the liberal totals than the liberal alternative, but higher than the conservative alternative, and/or (2) lower on the conservative totals than the conservative alternative, but higher than the liberal alternative. In the above generalized contexts, that means scoring between 10 and 14 on both columns.

SOS or win-win analysis is difficult to refute for the following reasons:

1. If the solution really does exceed the conservative alternative on the totals column that uses conservative goals and weights, then it is difficult for conservatives to object to the SOS solution.

2. Likewise, if the SOS solution exceeds the previous liberal best on liberal goals and weights, then it is difficult for liberals to object to it.

3. There are not very many neutrals in highly controversial choice or allocation situations. Even so, SOS solutions tend also to exceed the neutral best even though that is not one of the definitional elements of an SOS solution.

4. It is difficult to criticize the weights that are used because those weights are empirical rather than normative weights. The weights are not saying that a given goal has high or low importance. Instead, they are saying that conservatives or liberals tend to evaluate the goal under consideration as having high or low importance. That is an empirical statement about what is, rather than a normative statement about what ought to be.

5. It is usually difficult to criticize the relation scores. They are basically rank orders or relative direction on which there is likely to be consensus, as contrasted to more precise measures over which there may be much disagreement, although not necessarily disagreement which makes any difference in terms of the bottom line conclusions. For example, no two people are likely to say that the Washington Monument is smaller than the White House. At the same time, no two people are likely to agree through observation as to how many inches high the White House or the Washington Monument is. In other words, the more precision that is demanded, the more unnecessary disagreement is likely to occur.

6. If there is disagreement over the weights or the relation scores, the disagreement can possibly be easily resolved through the what-if analysis or threshold analysis. It is likely to show that regardless which position is right, the bottom line stays basically the same as to what is the optimum choice or allocation, because that choice or allocation is influenced by so many other things.

Win-win solutions have not been achieved very often in the past. The lack of occurrence is probably not due to the lack of potential existence of such solutions. Rather it is more due to the disputants thinking in terms of either compromises or something like war, strike, or litigation. What is needed is more thinking in terms of the possibility of arriving at super-optimum justice. Normally one thinks of self-fulfilling prophecies as being something undesirable, as where the prediction of failure helps cause failure. A self-fulfilling prophecy can, however, be something highly desirable. For example, predicting the possibility of a super-optimum solution helps facilitate such solutions.

This article has sought to clarify the concept of super-optimum justice with examples and general principles. Traditional philosophy has been concerned with having a just society, economy, and polity. Perhaps it might be more fruitful to think also in terms of the super-optimum society, economy, and polity. Win-win ideas can perhaps help stimulate the development of more applications and details of win-win policies toward the development of a win-win or super-optimum society, economy, and political systems.¹

NOTE:

1. For further details on win-win justice, see “Justice and Super-Optimizing Analysis” in Nagel, Stuart (1993). Legal Scholarship, Microcomputers, and Super-Optimizing Decision-Making. Westport: Greenwood-Quorum.. For earlier background, see “Public Policy Goals” in Nagel, Stuart (1984). Public Policy: Goals, Means, and Methods. New York: St. Martin’s. For more recent applications, see “The Ten Commandments and Win-Win Analysis” in Nagel, Stuart (1997). Super-Optimum Solutions and Win-Win Policy: Basic Concepts and Principles Westport: Greenwood-Quorum.