HL Fairness & UP PDF

Fairness and the UP Erasmus School for Philosophy Erasmus Institute for Philosophy and Economics (EIPE) Last time + This time Four SWFs............ UP = Ut + P1......... Kidney............... Lotteries for Kidney...... Bob should get a fair shake. Structure of this lecture..............................

Fairness and the UP Erasmus School for Philosophy Erasmus Institute for Philosophy and Economics (EIPE) Last time + This time Four SWFs............ UP = Ut + P1......... Kidney............... Lotteries for Kidney...... Bob should get a fair shake. Structure of this lecture................................. I. Different types of fairness What is fairness?............. From Aristotle to Broome....... The many faces of fairness...... Formal versus substantive fairness. Broad versus narrow fairness..... Global versus local fairness...... Objective versus subjective fairness Objective vs. subjective example.. Absolute vs. comparative fairness. Outcome vs. procedural fairness.. II. Broome’s theory of fairness Broome’s theory of fairness... One person Frio.......... Two person Frio.......... Do A because it is better?... Rights................. Consequentialist reasons..... Claims................. Examples of claims........ Incompleteness........... Taking stock.............................................................................................................................................................................................................................................................................................................................. 2 3 4 5 6 7 8............................................................................................................................................................................................................................................................................................................................................................................................................................................................................ 9 10 11 12 13 14 15 16 17 18 19............................................................................................................................................................................................................................................................................................................................................................................................................................................................................ 20 21 22 23 24 25 26 27 28 29 30 1 Two person Frio revisited. Two person Frio revisited. Lotteries............ Lifeboat revisited....... What to do in Lifeboat?.. Another Lifeboat....... Different Claim Strengths............................................................................................................................................................................................................................................................................................................................................................................. 31 32 33 34 35 36 37 III. Fairness and the UP Revisiting Kidney........... 1. Give up Consequentialism (P1) 2. Give up Welfarism (P2)..... 3. Re-describe the outcomes.... 3. Re-describe the outcomes.... 3. Re-describe the Outcomes............................................................................................................................................................................................................................................................................................. 38 39 40 41 42 43 44 Appendix 45 Summary 46...................................................................... 47 Literature 48...................................................................... 49 More Fairness?............................................................ 50 2 Last time + This time 2 / 50 Four SWFs We discussed 4 SWFs and their evaluation functions: Utilitarianism (Ut), Simple Egalitarianism (SE), Simple Prioritarianism (SP), MaxMin. Another name for the Utilitarian SWF is Sum-Ranking Welfarism. SE, SP, and MaxMin are all equality respecting but that they “care about equality in different ways”. Ut is not equality respecting. We illustrated these SWFs for 2 person utility profiles, x y z uA 4 6 6 uB 8 4 6 with (uA , uB ) a cardinal and comparable representation of well-being (We’ll continue using such utility profiles in this lecture) 3 / 50 UP = Ut + P1 Remember that the Utilitarian Principle (UP) is equivalent to: Sum-Ranking Welfarism (Ut): x is better than y iff the sum-total of well-being in x is greater than in y. P1 Consequentialism: You should perform the best action available to you. x uA 4 uB 8 Ut: x is just as good y z 6 6 4 6 z (and both are better than y). Ut + P1: both x and z are morally permissible (but y is not). 4 / 50 3 Kidney To introduce this week’s topic consider the following thought-experiment. Ann and Bob need a kidney transplant to survive but a single kidney is available. There are no important differences (in health, age or anything whatsoever) between Ann and Bob. Who should get the kidney? It is plausible to model Kidney as follows: uA uB oA 1 0 oB 0 1 ⎧ ⎪ ⎪ oA : Ann gets the kidney ⎨ ⎪ o : Bob gets the kidney ⎪ ⎩ B It then follows from Ut that oA is just as good as oB : Ut: oA ≈ oB because 1 + 0 = 0 + 1. Ut + P1: realizing either oA or oB is morally permissible. 5 / 50 Lotteries for Kidney Let us now compare two further actions (policies) for Kidney: (L1) Hold an equally weighted lottery to decide who gets the kidney. (L2) Give the kidney to Ann (“with probability 1”). uA uB oA 1 0 oB 0 1 ⎧ ⎪ ⎪ L1 ∶ oA with prob. 0.5, oB with prob. 0.5 ⎨ ⎪ ⎪ ⎩ L2 ∶ oA with prob. 1, oB with prob. 0 No matter whether you do L1 or L2: You’ll end up with an outcome with the same sum-total of well-being (1). And Ut only cares about the sum-total of well-being of outcomes. So Ut says: L1 is just as good as L2. And Ut + P1 says: both L1 and L2 are morally permissible. Now you might question these verdicts of the Utilitarian Principle... 6 / 50 4 Bob should get a fair shake (L1) Hold an equally weighted lottery to decide who gets the kidney. (L2) Give the kidney to Ann (“with probability 1”). Intuition: “L2 is not morally permissible, you should do L1!” Why? “Because L1 is fair and L2 is not” But what is fairness? Does fairness show that the Utilitarian Principle (UP) is false? 7 / 50 Structure of this lecture What is fairness? I. Different types of fairness II. Broome’s theory of fairness How to take fairness into account? III. Fairness and the UP 8 / 50 5 I. Different types of fairness 9 / 50 What is fairness? The importance of fairness becomes clear in any situation in which scarce goods such as health care resources or joint profits are to be divided. How to allocate these scarce goods in a fair way? What is fairness anyway? Philosophers often talk about fairness without giving a theory of fairness: The concept of fairness itself “is a central, but under-theorized, notion in moral and political philosophy” (Saunders 2010:41) “rarely if ever does one find a theoretical work of any sort devoted exclusively or at least largely to understanding what it means to be fair or unfair” (Carr 2000:3) Broome’s theory of fairness is one of the few exceptions. 10 / 50 From Aristotle to Broome Broome’s theory stresses that proportionality is key for fairness, a thought that is as old as Aristotle: Aristotelian formula. Fairness requires that equals should be treated equally, and unequals unequally, in proportion to relevant similarities and differences. Broomean formula. Fairness requires that claims are satisfied in proportion to their strength. Before we explain Broome’s theory in any detail, we will discuss the type of fairness it is a theory of. 11 / 50 6 The many faces of fairness As we will see, there are many types of fairness: Formal versus substantive fairness. Broad versus narrow fairness. Global (“social justice”) versus local fairness. Objective (“claims based”) versus subjective (“preference based”) fairness. Comparative versus absolute fairness. Outcome versus procedural fairness. Broome’s theory of fairness is: Substantive, narrow, local, objective, comparative, outcome and procedural 12 / 50 Formal versus substantive fairness Formal Fairness consists of interpreting and applying rules consistently, i.e. of applying the same rules impartially and equally to all. Soccer and Golf 1. A rule says that any soccer player betting against his own team will be banned from the sport. Dirk, Johnny and Kevin are caught betting against their teams and so get banned. Then Arjen is caught but he is so popular that he is not banned. This is (formally) unfair. 2. A rule prohibits Jewish people from joining a golfing club. This rule is applied consistently and impartially: no Jewish people are allowed to join. So this is formally fair. But it is substantively unfair, most people would say. So formal fairness is not sufficient for substantive fairness. Formal fairness is, however, necessary for substantive fairness. From now on we talk about substantive fairness. What then is substantive fairness? 13 / 50 7 Broad versus narrow fairness Fairness is often used with a very broad meaning. To say that a situation / decision / rule / action is unfair is to say that “it has some moral defect’.’ 1. The salary differences between men and women are unfair. 2. The decision to hire Piet but not Bao was unfair. 3. A lock-down is “unfair”, as other measures have better consequences. The moral defect in 3 is of a different kind than that in 1 and 2 The kind of moral defect in 1 and 2 (but not in 3) we call fairness. By doing so, we adopt a narrow conception of fairness. Hooker (2005) argues that it is theoretically and practically fruitful to have narrow conception of fairness. We concur. 14 / 50 Global versus local fairness Local fairness is about fairness in particular “local” situations: “Businesses are sensitive to issues of fairness in the salaries that they offer to their employees. Public agencies worry about fairness when they decide who has access to public housing; how much to charge for basic services such as water, electricity and public transport; who gets a kidney for transplantation; and who gets into a nursery school.”(Young 1994) Global fairness (“Social Justice”) looks at the overall way that a society distributes resources (and rights, duties, opportunities) across all places and times in that society. Young (1994) argues that, to resolve questions about local fairness, one does not need to first have a theory of global fairness. Soames (2019) says something much stronger: “[local] fairness is a pre-political social notion that must be understood before the basic principles for organizing the justice of a complex society can be established.” 15 / 50 8 Objective versus subjective fairness Subjective notions of fairness understand fairness in terms of preferences. Here, envy-freeness is the fundamental concept: Our approach to fair division is distinctive... in elevating the property of “envy-freeness”, and procedures that generate envy-free allocations [e.g. cut-and-choose], to a central place in the study of fair division. Roughly speaking, an envy-free division is one in which every person thinks he or she received the largest or most valuable portion of something—based on his or her own valuation—and hence does not envy anyone else. Brams & Taylor (1996) Objective notions of fairness understand fairness in terms of claims. Here, proportionality is the fundamental concept: This [Brams & Taylor] approach differs radically from that of the present book. The problem of fairness, as here conceived, is a matter of doing impartial justice rather than one of pleasing or satisfying the parties involved. The proportional co- ordination of shares with valid claims is the crux here. Rescher (2002) 16 / 50 Objective vs. subjective example Owing Money John owes €10 to Ann and €20 to Bob but has only €15 left. How should John divide his money? Objective fairness. Ann and Bob have (equally strong) claims to get their money back. In (€1, €2) their equally strong claims receive equal satisfaction (10 %). So (€1, €2) is comparatively fair. Only (€5, €10) is comparatively fair and efficient. Subjective fairness. Ann and Bob prefer to receive more money over less. Allocations in which they receive the same amount are envy-free. So (€1, €1) is envy-free / comparatively fair. Only (€7.5, €7.5) is comparatively fair and efficient. 17 / 50 9 Absolute vs. comparative fairness According to Broome, fairness is strictly comparative. E.g. for Owing Money, (€1, €2) is just as fair as (€5, €10). Claims ought to be satisfied. This requirement belongs to Broome’s general moral theory; it is not part of his theory of fairness. Piller (2017) According to others, fairness is not strictly comparative. Rather,it has a non-comparative, or absolute dimension, as well. “If Ann does not get all what she has a claim to, this is (absolutely) unfair to Ann” Absolute + comparative dimension: only (€5, €10) is fair. A recent theory of absolute and comparative fairness: Wintein & Heilmann, How to be absolutely fair part I: the fairness formula. Economics & Philosophy, 202x. Wintein & Heilmann, How to be absolutely fair part II: philosophy meets economics. Economics & Philosophy, 202x. 18 / 50 Outcome vs. procedural fairness Selling Cars The manager promises to give Ann a promotion when, within a year, she sells at least 50 cars. The manager makes the same promise to Bob. Both Ann and Bob sell over 50 cars within a year. Due to a budget-cut, only 1 promotion can be offered. The manager gives the promotion to Ann,as e.g.: Ann is friendlier / Ann sold 1 car more / Ann was hired earlier than Bob. Bob: that’s unfair! I sold ≥ 50 cars, so I was promised a promotion. Here, allocating the promotion via a lottery may alleviate Bob’s complaint. Here, a lottery is a fair procedure to allocate the promotion. Nevertheless, the resulting outcome will inevitably be unfair. “Sometimes a lottery is the fairest way of distributing a good, and my theory explains, better than any other theory I know, why this is so.” Broome (1990: 87) 19 / 50 10 II. Broome’s theory of fairness 20 / 50 Broome’s theory of fairness Broomean formula. Fairness requires that claims are satisfied in proportion to their strength. Broome’s theory is a theory about a notion of fairness that is: Substantive (not formal) Narrow (not broad) Local (not global) Objective (not subjective) Comparative (not absolute) Outcome and procedural. Let’s now articulate Broome’s theory in more detail, in particular let us explain the notion of a claim. 21 / 50 One person Frio John suffers from some deadly disease (call it Frio): Treatment I: will cure the Frio and leave him without chronic pain. Treatment II: will cure the Frio but leave him in chronic pain. Treatment I is better (for John). That’s because a life without chronic pain is better than a life with chronic pain, all other things being equal. This is not to deny that some people with chronic pain lead better lives than some people without chronic pain. 22 / 50 11 Two person Frio Bob lives with chronic pain whereas Ann does not. Ann and Bob both then contract the deadly Frio. There is only one (indivisible) unit of Frio medicine available which cures Frio and restores a patient to their pre-Frio state. Action A: give the medicine to Ann. Action B: give the medicine to Bob. The consequences of the actions: A: 1 person dies (Bob), 1 person lives a life without chronic pain (Ann). B: 1 person dies (Ann), 1 person lives a life in chronic pain (Bob). Action A is better than action B. Again, that’s because a life without chronic pain is better than a life with chronic pain, all other things being equal. 23 / 50 Do A because it is better? So for Two person Frio, Action A is better than action B. But intuitively there is some sense in which it is questionable / defective to give the medicine to Ann and not Bob. Broome argues that action A, i.e. giving the medicine to Ann, is unfair. To explain why A is unfair, Broome distinguishes between three different types of reason for giving a person (a share of a) scarce resource: 1. Rights 2. Consequentialist reasons 3. Claims. Let’s discuss these types of reasons in turn. 24 / 50 12 Rights One reason to give a person (a share of) the resource is because she has a right to the resource. “The owner of the medicine has a right to it and so should get it”. Rights determine outright what should be done: they trump all other reasons. Rights are not ”weighed up” against other reasons to see which reason is stronger. Two person Frio: neither Ann nor Bob has a right to the medicine. So rights can’t explain the sense in which action A is defective. 25 / 50 Consequentialist reasons One reason to give a person (a share of) the resource is because that would have the best consequences. Consequentialist reasons are ”weighed up” against one another to see which reason is strongest, i.e which reason, when acted upon, results in the best consequences. Two person Frio: It is overall best to give the medicine to Ann. Consequentialist reasons can’t explain the sense in which A is defective. 26 / 50 13 Claims One reason to give a person (a share of) the resource is because we owe it to that person to do so, i.e. because that person has a claim to it. Claims differ from consequentialist reasons. A politician demands a priority treatment at the hospital. If she does not get it, the budget of the hospital will be drastically diminished. There may very well be a consequentialist reason for giving the treatment to the politician, but she does not have a claim to it. Claims differ from rights, which “determine outright what to do”. Claims need to be “balanced against each other”, which is the task of fairness. But what exactly is / grounds a claim? Broome doesn’t tell us but “this might not matter because we understand talk of claims pre-theoretically” Piller (2017) 27 / 50 Examples of claims Owing Money John owes €10 to Ann, €20 to Bob and has €15 left. Investing Time Anna has spent one day a week on the joint project, whereas Beta has spent three days a week on it. After some time, the value of their project is 20K and Anna and Beta split apart. Dangerous mission Someone has to be sent on a mission that is so dangerous she will probably be killed. The people available are similar in all respects, except that one has special talents that make her more likely than others to carry out the mission well (but no more likely to survive). Owing Money. Ann’s claim to €10 is equally strong as Bob’s claim to €20. Investing Time. Anna and Beta both have a claim to receive a part of 20K, but Beta’s claim is, arguably, three times as strong. Dangerous mission. All available people (including the talented) have an equally strong claim to the good of not being sent. Indeed, these are different types of claims, as explained in Wintein & Heilmann, How to be absolutely fair part I. 28 / 50 14 Incompleteness Applying Broome’s theory of fairness will itself involve ethical judgements concerning what claims are claims to: Take the question whether the young should, other things being equal, get preferential treatment over the old when it comes to the distribution of scarce life-saving medical resources. The young would benefit more from these resources than the old would when we consider their effects in terms of life-years saved. Whether this difference in the size of benefits affects the strength of their claims crucially depends on what we understand their claims to be claims to. If people have a claim to a (reasonably) long life, the young will have a stronger claim. If, however, ill people simply have a claim to medical attention and treatment, differences in age will not amount to differences in claim strength. Piller (2017:218) Broome’s theory of fairness does not tell us what people have claims to and is, in that sense (deliberately) incomplete. But Broome’s theory is powerful: it helps us to structure moral debates. 29 / 50 Taking stock Broomean formula. Fairness requires that claims are satisfied in proportion to their strength. Broome’s theory is a theory about a notion of fairness that is: Substantive, Narrow, Local, Objective, Comparative, Outcome & Procedural. Claims are a particular type of reason for giving a resource to a person. Claims are reasons that are“owed to the person herself” Claims differ from consequentialist reasons and rights in “how they work”. Broome does not offer a theory of what people have claims to. Broome’s theory helps us to structure moral debates. Let us now illustrate this by revisiting Two person Frio. 30 / 50 15 Two person Frio revisited Bob lives with chronic pain whereas Ann does not. Ann and Bob both then contract the deadly Frio. There is only one (indivisible) unit of Frio medicine available which cures Frio and restores a patient to their pre-Frio state. Action A: give the medicine to Ann. Action B: give the medicine to Bob. Action C: give the medicine to no one. Ann and Bob have equally strong claims to the medicine as we owe it to each of them to save their lives. The consequences of the actions: A: 1 person dies (Bob), 1 person lives a life without chronic pain (Ann). B: 1 person dies (Ann), 1 person lives a life in chronic pain (Bob). C: two persons die. Via C, the equally strong claims receive equal satisfaction: no satisfaction! A is the best action (“does the most good”), but C is fairest. 31 / 50 Two person Frio revisited So there is a consequentialist (goodness) reason to do A. And there is a claims-based (fairness) reason to do C. So there is a sense in which it is defective to do A, even though A would be best overall: it is unfair to give the medicine to Ann and not Bob, because doing so doesn’t satisfy their claims equally. This solves our puzzle from a few slides ago, but it raises a new question: A is best and C is fairest but what should we do? (all things considered) 32 / 50 16 Lotteries “One of our aims must be to do as much good as possible, and it would surely be worth sacrificing some fairness to avoid the harm of allowing a person to die unnecessarily. So it would surely be right to save one of [Ann and Bob], even though this will inevitably lead to some unfairness.” Broome (1994: 39) “Even though unfairness is then inevitable, we can minimize it. We cannot treat both people equally, but we can at least give them an equal chance of being saved. We can decide whom to save by means of a lottery. An equal chance is not full equal treatment, but it is a second-best type of equality, and achieves a second-best type of fairness. In choosing to hold a lottery rather than let both people die, we are making some sacrifice of fairness for the sake of large gain in benefit”. Broome (1994: 39) So a lottery does not function as a tie-breaker, but as an allocation procedure that generates (some) fairness. For Two person Frio, a lottery strikes a proper balance between fairness and goodness: it is what we should do, i.e. it is the morally right thing to do. 33 / 50 Lifeboat revisited After a shipwreck, one person is stranded on some island a and five persons are stranded on some other island b. All six persons are strangers to you. You have a lifeboat which allows for a trip to one of the islands but, owing to a lack of fuel, you can only make one trip. When you use your boat to visit an island, you can rescue all persons which are stranded on that particular island. However, as an unavoidable consequence, all persons stranded on the other island will die. What should you do? Three possible actions: (1) Visit no island. ⎧ ⎪ ⎪Visit a (2) L = ⎨ ⎪ ⎪ ⎩Visit b with probability with probability 1 2 1 2 (3) Visit island b. (1) is fairer than (2), which is fairer than (3). (3) is better than (2), which is better than (1). 34 / 50 17 What to do in Lifeboat? Here is what Broome (1998: 956) recommends for Lifeboat: But as I said, fairness is not everything. Fairness requires tossing a coin, but just as I think the fairness of saving no one is outweighed by the badness of the result, so I think the fairness of tossing a coin is outweighed by the expected badness of the result. Tossing a coin will lead you to save three lives on average (the expectation of lives saved is three), whereas you could save five for sure. Two lives are worth some unfairness, I should say. Therefore, I think you should save the five without more ado. More generally, the requirement of fairness will typically conflict with the requirement of “doing the most good”. How should we resolve this conflict between fairness and doing the most good? [... ] I cannot resolve the conflict, but I can try to contribute toward a theoretical understanding of it. Broome (1994: 37) 35 / 50 Another Lifeboat After a shipwreck, 10.000 persons are stranded on some island a and 10.001 persons are stranded on some other island b. All persons are strangers to you. You have a lifeboat which allows for a trip to one of the islands but, owing to a lack of fuel, you can only make one trip. When you use your boat to visit an island, you can rescue all persons which are stranded on that particular island. However, as an unavoidable consequence, all persons stranded on the other island will die. What should you do? Three possible actions: (1) Visit no island. ⎧ ⎪ ⎪Visit a (2) L = ⎨ ⎪ ⎪ ⎩Visit b with pr. 12 with pr. 21 (3) Visit island b. Is the fairness of tossing a coin still outweighed by the expected badness of the outcome in this case? Broome’s theory does not tell us how to weigh fairness against goodness (which is another sense in which the theory is incomplete). 36 / 50 18 Different Claim Strengths In Two person Frio and in Lifeboat, all claims are equally strong. What if claims differ in strength? Suppose that there is available some limited quantity of medicine and that this medicine cannot be divided without rendering it ineffective. Suppose your claim on the medicine comes from the fact that that you need it to save your life, and my claim on it comes from the fact that I need it to save my little finger. Suppose an average life is something like a thousand times more important than a little finger. What to do? (I) Weighted lottery (1000 lottery tickets vs 1). (II) Let the strongest claim win. In the literature, both (I) and (II) argued for. There is disagreement about why you should do (I) / (II). We won’t have time to discuss this in any detail. 37 / 50 19 III. Fairness and the UP 38 / 50 Revisiting Kidney Two lotteries for Kidney: L1. Hold an equally weighted lottery to decide who gets the kidney. L2. Give the kidney to Ann (“with probability 1”). Intuition: you should do L1 (L2 is not morally permissible) as L1 is fairer. We will assume that Intuition is correct. Also, we assume that fairness is described by Broome’s theory. How then, to reconcile fairness with the UP) (which contradicts Intuition? We discuss three possible answers: 1. Give up Consequentialism (P1) (How we analysed Two-person Frio) 2. Give up Welfarism (P2) 3. Re-describe (the outcomes of) L1 and L2. (Broome’s official answer!) Broome: fairness is compatible with both Intuition and the UP. 39 / 50 20 1. Give up Consequentialism (P1) uA uB oA 1 0 oB 0 1 ⎧ ⎪ ⎪ L1 ∶ oA with prob. 0.5, oB with prob. 0.5 ⎨ ⎪ ⎪ ⎩ L2 ∶ oA with prob. 1, oB with prob. 0 Intuition: You should do L1 (and not L2) because L1 is fairer. You might argue that Intuition is true, because: Consequentialism (P1) is false: what you should do is not (only) determined by what is best: “the right is not determined in terms of the good” Indeed, what you should do is determined both by goodness and by fairness. You might think that principle P2, P3, P4 and P5 are all true. It then follows that L1 is just as good as L2. It does not follow, however, that L2 is morally permissible, because P1 is false. 40 / 50 2. Give up Welfarism (P2) uA uB oA 1 0 oB 0 1 ⎧ ⎪ ⎪ L1 ∶ oA with prob. 0.5, oB with prob. 0.5 ⎨ ⎪ L2 ∶ oA with prob. 1, oB with prob. 0 ⎪ ⎩ Intuition: You should do L1 (and not L2) because L1 is fairer. You might argue that Intuition is true, because: Welfarism (P2) is false: how good an alternative is, is not determined solely by well-being. Instead, it is determined both by well-being and by fairness. You might think that principle P1, P3, P4 and P5 are all true. And that you should do L1 (and not L2) because L1 is better than L2. E.g. you might think that L1 is better than L2, because: euA (L1) + euB (L1) + F (L1) > euA (L2) + euB (L2) + F (L2) with F (⋅) some measure of how fairness contributes to goodness. 41 / 50 21 3. Re-describe the outcomes uA uB oA 1 0 oB 0 1 ⎧ ⎪ ⎪ L1 ∶ oA with prob. 0.5, oB with prob. 0.5 ⎨ ⎪ L2 ∶ oA with prob. 1, oB with prob. 0 ⎪ ⎩ Intuition: You should do L1 (and not L2) because L1 is fairer. You might argue that Intuition is true, because: The outcomes oA , oB are under-described: If Ann gets the kidney via L1 she has received a fairly allocated kidney. If Ann gets the kidney via L2 she has not received a fairly allocated kidney. But oA : Ann gets the Kidney does not distinguish between these cases. Outcomes should be re-described as follows: o1A : Ann receives a fairly allocated kidney (via L1). o2A : Ann receives a kidney, but not one that is fairly allocated. (via L2). 42 / 50 22 3. Re-describe the outcomes Broome argues that: (i) The outcomes of the lotteries should be re-described to account for fairness. (ii) (Un)fairness itself affects well-being, so that Kidney’s lotteries should be modelled along the following lines: uA uB o1A 1 0.1 o1B 0.1 1 o2A 0.9 0 o2B 0 0.9 ⎧ ⎪ ⎪ L1 ∶ o1A with prob. 0.5, o1B with prob. 0.5 ⎨ ⎪ L2 ∶ o2A with prob. 1, o2B with prob. 0 ⎪ ⎩ L1 results in an outcome with a sum-total of well-being of 1.10 L2 results in an outcome with a sum-total of well-being of 0.9 So Ut says: L1 is better than L2. And Ut + P1 says: you should do L1, L2 is not morally permissible. Fairness is compatible with the UP and our Intuition that we should do L1. Or so Broome argues. However... 43 / 50 23 3. Re-describe the Outcomes The idea that fairness is a part of well-being is highly controversial. It doesn’t seem consistent with any of the theories of well-being of L4A: Preference satisfaction / Hedonism / Nussbaum’s Objective list theory. In order for re-describe the outcomes to be successful: Even if Ann and Bob will never know how the kidney is allocated, it generates more well-being when the kidney is allocated fairly rather than unfairly! If re-describe the outcomes is unsuccessful, then UP is false. Indeed, when P1 or P2 is false, UP is false as well. Next week: more reasons to think that UP is false. 44 / 50 Appendix 45 / 50 Summary 46 / 50 There are various types of fairness, Broome’s notion of fairness is: Substantive, Narrow, Local, Objective, Comparative, Outcome & Procedural. Broome: Fairness requires that claims are satisfied in proportion to their strength. Claims are a particular type of reason for giving a resource to a person, that are“owed to the person herself” and that differ from consequentialist reasons and rights in “how they work”. In Two person Frio, allocating the good via a lottery yields a second-best type of fairness, strikes a proper balance between fairness and goodness and is what we should do. In other cases (e.g. Lifeboat) the fairness and goodness may strike a different balance. How to take fairness into account? Give up Consequentialism / Give up Welfarism / Re-describe the outcomes. 47 / 50 24 Literature 48 / 50 1. Broome, J. 1990. Fairness. Proceedings of the Aristotelian Society 91: 87-101. 2. Broome, J. 1994. Fairness versus Doing the Most Good. The Hastings Center Report 24(4): 36-39. 3. Broome, J. 1998. Kamm on Fairness Philosophy and Phenomenological Research 58(4): 955-961. 4. Piller, C. 2017. Treating Broome Fairly. Utilitas 29(2): 214-238. 5. Wintein, S. & Heilmann, C. 2021. Fairness and Fair Division. In: Heilmann, C. & J. Reiss (eds.) The Routledge Handbook of Philosophy of Economics. 255-267. (1) is the original exposition of Broome’s theory of fairness. (2) is a very short description of Broome’s theory and its relation to QALYS. (3) applies Broome’s theory to Lifeboat and contrast it with Kamm’s theory. (4) clearly explains Broome’s theory and points out that various objections that have been raised to it are based on misunderstandings of the theory. (5) briefly sketches Broome’s theory and points out relations with economic theories of fair division. The first part of (5) was used to structure this lecture. 49 / 50 More Fairness? Together with C. Heilmann (EIPE) I run the Fairness Project: https://fairness-research.org/about/ 50 / 50 25