Fairness in Resource Allocation
Michael A. Gillette, Ph.D.

There can no longer be any doubt that we are operating in a healthcare environment in which we must make difficult decisions regarding the allocation of scarce resources. These decisions can be made either on the micro-allocation level where the competing needs of individual consumers are considered, or on the macro-allocation level where organizational budgetary priorities are set. It is with some relief that I now discuss this topic in this forum, because it signals the fact that allocation decisions – which have actually been made in this country for many years – will now be made intentionally and carefully rather than by accident and without moral ownership.

When dealing with the question of allocating scarce resources, our first goal should be to identify the conceptual framework with which we are working. There is no doubt that many practical problems must be identified and solved before a workable system of allocation can be developed. Nevertheless, the goals of our allocation efforts must be fully understood in order to measure the ethical acceptability of any particular outcome. To begin discussion on this topic, I would like to introduce the “Four E’s of Allocation”. The Four E’s are effectiveness, efficiency, equality and equity. While many of us find ourselves attracted to one or the other of these values, we rarely recognize the values that we are operating under and we fail to openly discuss the values that we have adopted. This fact leads to an ethically untenable position since, without a better understanding of our core values, we are unable to defend our actions when challenged.

The table below represents the Four E’s of allocation in numerical form. This table displays four possible worlds in which we could live. More exactly, it represents the outcomes of four distinct distributional schemes. A and B on the left side of the table represent the two individuals or groups of individuals that stand to benefit when we select a particular distribution of goods. These goods might include financial support, staff time, consumer autonomy, consumer security, etc… The numbers under alpha, beta, gamma and delta indicate the value of the outcome of each of those distributions to A and B, on whatever grounds you choose to make an outcome assessment. In short, A and B are the parties involved in our simplified thought experiment, and alpha through delta are the outcomes of the possible allocations that we might make.

The Allocation of Resources
Effectiveness, Efficiency, Equity, Equality

In world alpha, the party making the allocation decisions has clearly adopted the value of effectiveness. For our purposes, effectiveness can be defined as the situation in which we make the greatest possible gain for one particular individual. According to this value, we have succeeded in making an ethical allocation choice because the person who gains most by our actions gains the greatest amount that he/she possibly could (A’s gain of 1000 is greater than any other value on the chart).

It is common for staff to find themselves thinking in terms of effectiveness when they have a clearly identified case load and are likely to develop loyalties to specific consumers. While this may feel comfortable for staff members who face their own consumers on a regular basis, many would complain that this approach to allocation, which generates a great disparity in outcomes for consumers, is unfair. Many would argue that we must look at the benefits that our organization creates across its consumers. We must strive to maximize the benefit that we create in totality.

If this line of reasoning sounds impressive to any particular reader, then you are likely to prefer the concept of efficiency to that of effectiveness. Such a reader would prefer world beta to that of world alpha. In alpha we made the greatest difference to consumer A, but consumer B was left with little. If we examine world beta, however, we see that the combination of what A receives and what B receives (1,025 points) is greater than the combination of benefit that A and B received in alpha (1,010 points). In fact, a quick examination of the table indicates that world beta has the greatest overall distribution of benefit.

Although there is no doubt that beta is the most efficient of all our possible distribution schemes, since the overall output is the greatest (and we can assume identical input across options), many would still argue that the disparity between A’s welfare and B’s welfare is unacceptable. Although A and B combined do better in beta than under any other distributional system, the disparity between them is still great. Many would argue that this situation seems unfair and that equality must be considered an important value in making allocation decisions.

If equality truly matters, then world gamma should be preferred. Under that system of distribution, A and B benefit equally. It should become immediately clear, however, that world gamma is not the preferable world in which to live. Although A and B receive equally valuable outcomes under gamma, those equal outcomes are equally unfortunate. It makes no sense for either A or B to prefer to live in world gamma, since they both do worst in that world. Any other choice would be preferable for both A and B. If equality matters ethically, its value must be tempered by some other factor.

Delta represents a distribution in which equity is preferred over equality. In world delta A does not benefit as much as he would in the other worlds, but B does much better than he would in any of the other worlds. It is true that delta is more equal than alpha or beta, but it is less equal than gamma and less efficient than either alpha or beta (total output of 800). Nevertheless, if this table represents all of the real options that are available, it would be irrational for B to prefer any situation over delta. Delta is the fairest of all the distributions listed because the party that loses under that scheme loses least. The fairest position is the one under which the least advantaged member of the group is better off than the least advantaged member of the group under any other rationing system.

The rationality of this choice should be clear. If B prefers any system other than delta, he is effectively saying that he would rather do worse, just so A does not do better. This position is spiteful, however. We should ask, as did John Rawls in A Theory of Justice, what a rational but unbiased person would choose. If you did not know whether you were A or B, but you did know that you were one or the other, and your choice of distributions would be permanent once it was made regardless of whether you turn out to be A or B, which world would it be rational choose? I believe, as does Rawls, that delta is the obvious unbiased choice.

We can conclude several things from this discussion. First, we can conclude that a fair approach to rationing is one that works to maximize benefit to the least advantage member of the group under consideration. Second, we can conclude that once minimum standards are satisfied for everyone, additional resources should be used to improve the situation of those least advantaged. This has the counter-intuitive result of forcing us to look more closely at our failures in providing services rather than our successes. This approach forces us to forego possibly great gains for a few, in order to verify that no one falls below the safety net. This approach requires that we consider minimum standards of care before we pursue best practice. This approach is most equitable, however, and therefore the most fair. This entire discussion supposes, however, that the shortage is real and that no other resources are available to help produce greater benefit for everyone involved. My conclusions only hold true if there really is no world epsilon out there where A and B can both do better.