On the Use of the Term "Self-Interest" in Economics

I've been repeatedly embroiled in an argument for the last few weeks over the term "self-interest" as it is used in economic discussion, and I wanted to hammer out my position once and for all so that I don't have to keep trying to start from the beginning. Here's the deal. I am told that within the discipline of economics, what it means to say that a person "acted in her own self-interest" is that a person "acted according to her own interests." The idea here is that all action demonstrates preference, and that this necessarily means that the actor preferred the action that was taken to all other actions. So if I jump on a grenade in order to save my friends, what I have demonstrated is that I preferred to jump on the grenade over all other alternatives that I considered, and it's fair to say that I wanted to jump on the grenade; that out of all available alternatives, the one I consider the best is the one where I jump on the grenade so that my friends live. I'm down with that.

When I jump on the grenade because I want to save my friends, I take it to be uncontroversial that I do so according to my own interests. How could it be otherwise? And if what we mean by "self-interest" is simply that I act according to my own interests, then yes, my jumping on the grenade is self-interested.

But when presented with the claim that jumping on the grenade is a self-interested behavior, the average person tends to become perplexed. It's only after a thorough explanation of the "economic" meaning of the term that it becomes clear how this could be the case. Why does this happen? The reason, I contend, is that economists mean something completely different by the term "self-interested" than lay people do. This, I will argue, is a problem, and should be remedied in order to prevent completely unnecessary confusion and error.

Let me explain. In talking about any interest or preferred scenario, there must be a subject and an object. The subject, generally speaking, is the person who has the interest or the preference. So if we're talking about my preference for eating an apple, the subject is me. It is I who prefers the apple, and the preference for the apple is incoherent without the fact that the preference is my preference. The object of the preference, on the other hand, is the end which the subject is seeking to promote. In our example, I prefer the apple, but the object of my preference is not simply the apple: I don't value the apple for itself. I want to eat the apple. The object of my preference, then, is something along the lines of my having eaten the apple (perhaps we might say that I want "the experience" of eating the apple, or "the happiness" produced by my eating the apple; the exact way we phrase this is not critical).

The critical thing to note here is that the economists' definition of "self-interest" simply refers to the idea that interests are subjective: the subject of all interests is the interested individual. It is my understanding, however, that when lay people use the term "self-interest," what they have in mind is, minimally, that the object of the preference has something to do with the interested individual. So if my sister were sick, I might go get her some medicine. To say that my getting the medicine is "self-interested" would mean, to the lay person, that I get the medicine in order to promote some self-directed end. That is, I get the medicine because, perhaps, I am happier when my sister is not sick, or my sister is irritating when she's sick, or there's a cute pharmacist who will think I'm sweet for taking care of my sick sister. The lay-person, then, would call "non-self-interested" or "selfless" an interest with an object which does not directly involve the actor. So I act selflessly if the reason I go get the medicine is that I value my sister's health for its own sake, and am willing to take on the costs necessary to promote her health.

Note that this lay definition of self-interest is not incoherent or contradictory. And note also that the "selfless" act identified by the lay definition is labeled as "self-interested" by the economist definition. Indeed, the notion of "selflessness," as identified by the lay definition, is defined out of existence by the economist definition. Because the economist identifies as "self-interested" all actions where the subject is the actor, and because all actions demonstrate an interest on the part of the actor, it becomes clear that there can be no such thing as a "non-self-interested" or "selfless" action.

A number of problems immediately present themselves. The first problem is that the economist definition completely eliminates what I take to be an extremely useful distinction between "self-interested" and "selfless" actions, which is captured very well in the lay definition, without providing an adequate substitute. One might object that the term "selfish" captures the layman's "self-interested," but to most people, the term "selfish" is emotionally charged with negative connotations. Observe the struggles of the Objectivists to try to divorce this emotional
connotation from the term! By contrast, the layman's "self-interest" is relatively neutral and already conveys the sort of thing that the economist would be trying to bend "selfish" into meaning. Further, the economist would then need a new word for the layperson's "selfish"!

Another reason that the fundamental difference between the lay person's and the economist's definition is undesirable is that the economist's definition of "self-interested" means exactly the same thing as the lay person's "interested." Because all interests are subjective, and the "self" in "self-interested" refers only to this fact, the term becomes redundant. The only thing that could conceivably be added by using the term "self-interest" would be if the addition of the "self" served to remind people that preferences are subjective. But as we have discussed, the term "self-interested" already means something, and it has nothing to do with subjectivity. If anything, the use of the term crowds out more useful terminology like "subjectively-interested."

Yet another problem with the economists' definition is that now we have a situation where the technical definition of the term "self-interested" is fundamentally different from the normally accepted definition of the word. That means that in order to actually communicate their points to lay people, economists will need to first make clear what they mean by self-interested, and ensure that their audience keeps this definition firmly in mind so as to avoid drawing bad conclusions. This also creates a systematic likelihood that people will be misled by economists who fail to properly emphasize their use of the redefined term. Nowhere is this problem more apparent than in the field of Public Choice economics. We might imagine an economist going before a crowd of lay persons and announcing that "The problem with governments is that they are run by self-interested people." We might imagine that what the economist means here is that politicians act according to their own preferences, and do not magically take on "society's" preferences when they are elected to office. They are, after all, human! And this would be a good and important point. But upon hearing the economist say that politicians are self-interested, a number of lay people might interpret the economist as making the argument that politicians are "in it for themselves" and are simply involved in politics in order to accrue benefits for themselves, regardless of whether others are harmed in the process. If it's true that the economists' use of the term "self-interest" does not offer any new or important insight into anything, as I argued above, it's unclear why we wouldn't want to simply avoid this problem altogether.

The final problem with the economists' use of the term "self-interested" is that economists themselves may end up misusing the term and reverting to the normal definition without noticing. Remember, economists are lay persons before they are economists, and have generally grown up with a meaning of the term "self-interested" which is very different from the meaning they've been trained to adopt in their profession. As a result, you end up with phenomena like economists saying things along the lines of "Because all actions are self-interested, it's clear that the reason you jump on the grenade is because you would be miserable if you didn't, and you expect that the misery would be way worse than dying." And I assure you, having heard that point made today, the risk of this sort of thing occurring is very real.

So in conclusion, I say that economists should quit their ridiculousness and give us back "self-interest." Their definition takes away a useful distinction which is captured by the normal meaning of the word, doesn't explain anything new, and doesn't accomplish anything except confusing everyone, including the economists themselves.

Cap and Trade vs. the Carbon Tax

So I've been addressing the issue of anthropogenic climate change for some time now, and I haven't said much in the way of addressing specific policy proposals. But I was just given a delightful present by one of my fellow FEE associates: a copy of the American Institute for Economic Research's latest Economic Education Bulletin, entitled "The Global Warming Debate: Science, Economics, and Policy." I didn't read the whole thing, but my favorite part was definitely when William R. Cotton, a professor of atmospheric science at Colorado State, closed his completely science-oriented essay, "Summary View of Climate Change," with:

There are strong indications that our global climate is warming. But the question is, is the warming due to anthropogenic greenhouse gases, or is it due to some other forcing mechanisms (or their transient absence) and natural variability. As human population on Earth continues to increase, the chances of human-induced changes in climate due to greenhouse gases, aerosol pollution, or alterations in land use become increasingly likely. Thus, rather than consider climate engineering, we should devise methods of encouraging the reduction of population growth through economic and quality-of-life incentives.

Period, end of conversation. No comment on that gem anywhere else in the entire essay. Who's got two thumbs and loves it? This guy.

But anyway, that's not the point. Later in the publication was an essay by Kenneth P. Green, a resident scholar at the American Enterprise Institute, where it was argued that a carbon tax is superior to a cap-and-trade system. I bounced between frustration, amusement, and glee as I read it, and felt an immediate need to comment. Not because Green did a bad job--he did just fine--but because he was guilty of something which is very common among people who discuss climate change: he discussed the possible "solutions" to climate change without addressing the reasons that a policy was to be implemented in the first place, and how the different solutions worked to address those reasons. His argument for a tax scheme over a cap-and-trade scheme was simply that a tax scheme could achieve the same goals, but with better economic side-effects and less potential for failure. Fine, I'll even grant it. But taxes and caps are fundamentally different policies, which only make even a little sense when confronted by specific sorts of problems.

I should explain what I mean. I've discussed elsewhere the idea that in order to make any sense from an ethical point of view, pollution taxes need to be based on the idea that an individual is justified in polluting if and only if she pays compensation to her victims for any damage done to them. That idea is controversial, but for our purposes we don't need to address that controversy. The point is only that even if we accept that idea as true, there are still only certain kinds of instances in which the injustice of pollution can legitimately be dealt with through a tax on pollution. The paradigm cases are those instances in which the damage caused by pollution is directly proportional to the amount of pollution that there is, so that the tax becomes the "price" of compensating the victims of one's actions for the costs one imposes upon them.

Cap and trade schemes, on the other hand, are built for an entirely different kind of problem. In a paradigm cap and trade situation, there is a threshold level of pollution with which policymakers are concerned, and at the threshold, a certain amount of damage is anticipated. The cap and trade scheme accordingly sets the cap at the relevant amount of pollution, and then distributes "shares" of the "environmental space" below that threshold in some way (e.g., auction, grandfathering system...). Because the allocations may be economically inefficient for whatever reason, the shares can then be traded in accordance with the wishes of their owners in order to ensure that the right to pollute is distributed to those individuals who are willing to pay the most for it (note that the normal objections to the "willingness to pay" criterion are avoided by passing the buck to the distribution process, which of course must be justified separately).

The point I want to make here is that global climate change is a very different phenomenon than the sorts of phenomena for which either of these policies is built to provide a solution. As noted elsewhere, climate change is an emergent problem. That is, climate change is not the result of any individual's actions, but rather is the consequence of many individuals acting separately, so that no individual can reasonably be said to have been able to prevent climate change from occurring, and no individual could have caused climate change singlehandedly. Accordingly, it does not make sense to talk about the consequences of climate change in terms of marginal contributions. The amount of damage caused by climate change will not likely change recognizeably with an additional increment of CO2 (or any other forcing agent), so it's not reasonable to try to put a price on how much damage "a unit of climate forcing" (expressed, perhaps, in terms of GWP, or Global Warming Potential, as defined by the IPCC?) causes.

A tax on contributions to climate change, therefore, seems like a policy which would require a bit of shoehorning. Individuals paying the tax would not be paying the "social cost" of their particular contribution, taken in isolation, because that would be basically zero. They would need to be charged for their "portion" of the total amount of damage done by climate change. So what policymakers would need to do would be to determine the total amount of damage which would be done at the equilibrium price for pollution permits, and then sell the permits at that price. The problem then becomes one of economic calculation. It could be done to some degree, but it would be inherently imprecise. And remember: the end result needs to be that the victims get compensated, so the government would have to go into its own pockets (that is to say, the pockets of its treasury or, more realistically, the pockets of its Federal Reserve printing press) to take care of the balance if it aimed low. And as my wonderful economist friends would point out, there would be a considerable incentive to aim high, creating a surplus revenue stream for the government which would almost certainly not be returned. So the tax is doable, kind of, but the problem is not the kind of thing that the tax is designed for. It's just that you can use the tax to accomplish the end goal if you want.

The cap and trade system is a little harder to adapt to the task, but there are a number of ways that the idea can be useful. First, there is a level to which we could collectively exert a forcing on the climate system without producing objectionable consequences. This level of climate forcing is a threshold which could be amenable to a soft cap and trade scheme (soft like the baseball salary cap). In this kind of policy, the cap would be set at the level of forcing which would produce no negative consequences, and this "environmental space" would be allocated somehow (or, if people find this to be a bad idea, we would simply say that these shares should be allocated in proportion to one's contribution to climate change, so that the soft cap has no effect). People not receiving these shares, or polluting in excess of their shares, would be filling environmental space which represented something like "harmful social emissions". Because these emissions would not be legitimated by the soft cap, they would be the ones which would be subject to the obligation to compensate the victims (again, if the soft cap isn't being used, as mentioned above, it would just be that everyone would have to participate in compensating the victims).

Here a potential for another cap would become apparent: We might imagine that policymakers would decide on a level of pollution (corresponding to some amount of total damage) which was determined to be "socially desirable" somehow. Perhaps, using the same reasoning involved in the tax scheme discussed above, the policymakers would arrive at the level of pollution which would clear the market if everyone paid some price for it. Or perhaps the policymakers would identify a level of pollution beyond which unacceptable results would occur, and the cap would be set there. In any case, you would then have to set a cap and allocate the shares. So again, the policy could be made to work. But the problems are simply that it's difficult to identify a level of "unacceptable" pollution, it's just as difficult to identify a market clearing price in this scheme as it is with the tax (assuming that the shares are auctioned, of course), and any other way of running the scheme is sure to carry either difficulties of its own, or charges of arbitrariness which would sever the connection between the problem and the solution.

So ultimately, what we're faced with is a situation in which the only two policy suggestions that are on the table are not particularly well suited to the task of "solving" the problems arising from climate change (and I haven't even begun to address the question of how the compensation process would even work, or whether compensation could make climate change legitimate!), and the only way to make either of them work is to basically stretch and contort them until they are made to do the job acceptably. Doing so, it will be noted, requires in both cases that government decision-makers possess knowledge and foresight which they almost certainly do not have, and even then it's unclear that the policies would work properly.

Obviously, there's a lot more to say about this. I just wanted to get some preliminary thoughts down, and I think this was a good start.

Another Double Standard Between Governments and Individuals?

So today was my first day at the Foundation for Economic Education, where I'll be interning over the summer, and I've already had some excellent debates; this is going to be a fantastic experience. Everyone seems really passionate and interesting, and I'm sure I'm going to learn a lot from everyone. I wanted to put one of the more controversial debate topics on my blog as a record, and to get the idea out to a wider audience. I've been toying around with the idea for a few days; I'm really curious to hear what other people think.

The idea is this: If we recognize private entities' claims to property titles as legitimate, even when they have a known history of violence and illegitimacy, then it's difficult to argue that currently existing governments are illegitimate for property rights-based reasons. Governments claim that we live in their territory, and their claims have roots that go back many generations. To claim that a government is not justified in enforcing rules in its territory is, effectively, to claim that the government is not the legitimate owner of that territory. But saying that, it seems to me, makes it very difficult to consistently argue that many (most, if not all) private property titles are legitimately held.

We had a bit of fun with this one at dinner, and I'm not completely sure what I think of it. Of course, everyone else at the table was not too comfortable with the idea, and it made for some lively debate. But nevertheless, I figured I'd post it here. Feel free to leave any comments; I'll be interested to hear what people think about this.

The Responsibility Principle vs. Breach of Duty

So I stumbled upon a really jarring debate today. I'm sort of puzzled that I haven't already heard of this issue, and am suspicious that someone might just be able to explain to me why there isn't any problem, and I'm just confused. But in any case, here's the issue.

It seems that in our current legal system, in order to establish that someone owes you damages to compensate you for a tort, you need to show that they have breached a duty that they owed to you. If it is determined that they did nothing wrong in harming you, then the idea is that they don't owe you anything.

But on the other hand, there's this, care of Joel Feinberg:

Suppose that you are on a backpacking trip in the high mountain country when an unanticipated blizzard strikes the area with such ferocity that your life is imperiled. Fortunately, you stumble upon an unoccupied cabin, locked and boarded up for the winter, clearly somebody else’s private property. You smash in a window, enter, and huddle in a corner for three days until the storm abates. During this period you help yourself to your unknown benefactor’s food supply and burn his wooden furniture in the fireplace to keep warm. Surely you are justified in doing all these things, and yet you have infringed the clear rights of another person.

I agree that the hiker is justified in his actions. But as Judith Thomson points out, it seems true that in this case, the hiker would also be obligated to compensate the owner of the cabin for the damage. This is in line with a principle central to the doctrine of Strict Liability, called the Responsibility Principle. Talbot Paige phrased the principle like this: "When A's actions impose costs on B, A should be made responsible, by paying those costs." It sort of does seem like this is why the hiker should have to compensate the cabin owner. Even though the hiker didn't do anything wrong, he still imposed a cost on the cabin owner, and he should have to pay that cost.

So it seems like I'm rejecting the "duty of care" standard. But on the other hand, I feel like there are some situations in which Strict Liability is, well, too strict. It seems to me that the concept of negligence (as distinct from something like "mere harming") is not completely without value: I find it an attractive notion that in situations where a person does nothing wrong, they should not be subject to the coercive pressure of others (through being held to account for something by a court--here I obviously don't mean "coercive" to imply that there's anything objectionable about holding people accountable through courts).

I definitely need to think about this some more; any thoughts or suggestions would be very much appreciated!

Enforcing the Attitude of Respect for Nature

So I wrote this for my environmental ethics class, and I figured I'd post it. But I just want to make clear up front that I don't agree with Taylor's argument for the attitude of respect for nature. The paper takes its correctness as a premise, and I want to make sure it's understood that I believe that premise to be false. Nevertheless, I think it's a fascinating point of view, and I think this exercise has been valuable for me in terms of thinking about rights and how they're supposed to work. Hopefully others can find some value in it too.

--------------------------------------------------------------------------------------

In his book, Respect for Nature, Paul Taylor offers an account of a disposition which he calls the “attitude of respect for nature.”[1] This disposition, Taylor explains, involves the recognition of the inherent worth of non-human “nature,” which leads to the treatment of nature with a proper degree of respect.[2] Proper respect, he argues, involves adherence to certain rules[3] which should not be broken in the absence of morally significant reasons. These rules include a duty to refrain from harming other organisms, from interfering in their activities, and from deceiving them or violating their trust, as well as a duty to make restitution when one does wrong.[4] For Taylor, the attitude of respect for nature is a virtue; it is view about how good people will act,[5] and for what reasons they do so.[6] For the sake of our discussion, we will make the assumption that Taylor’s view is objectively sound (a claim which Taylor does not make[7]), and the attitude of respect for nature is in fact virtuous.

Because Taylor’s view includes a mechanism by which individuals would voluntarily make restitution for any transgressions, it seems that if everyone were completely virtuous, there would never be any need to coercively enforce the mandates of justice; Taylor’s virtuous people would take care of any problems on their own. But it will immediately be noted that not all people are virtuous in Taylor’s sense. Many fail to hold the attitude of respect for nature. Because of this, they will likely do things that are inconsistent with this attitude, and which therefore represent iniquities. Accordingly, we might ask whether others would be justified in coercively interfering with these individuals’ actions in order to ensure that individuals who are not themselves virtuous nevertheless behave as if they were.

Generally, the simplest way to justify coercion is to cite the violation of some right held be the victim. As Taylor points out, rights represent things to which we are entitled,[8] so that if we are wrongfully deprived of them, our rights are violated.[9] Given that we are entitled to the objects of our rights, it seems eminently plausible that we would be justified in using force in response to the violation of our rights. If we could show that the sort of unvirtuous actions identified by the attitude of respect for nature represented rights violations, then we would have a good reason for believing that coercive enforcement might be justifiable.

Complicating matters, however, is the fact that Taylor offers two different accounts of rights, and these two views produce opposite conclusions about whether the behaviors in question violate rights.[10] In order to properly appraise the justifiability of coercion, we will need to decide between these views. According to Taylor’s first account, there are several necessary components of rights which plants and non-human animals cannot fulfill, making it impossible to conceive of them as having rights.[11] First, Taylor argues that to have a right, it must be possible for us to conceive of the rights-holder asserting the moral legitimacy of the claim represented by a right. Because they cannot think in terms of moral legitimacy, plants and non-human animals fail this test.[12] Second, he claims that we must be able to conceive of the holder of a right being able to think of herself as being inherently worthy of that right. Plants and non-human animals fail in this criterion as well.[13] Third, Taylor contends that we must be able to conceive of a rights-holder as being able to choose whether or not to exercise his right. Because plants and non-human animals cannot make this sort of choice, they fail here as well.[14] Finally, Taylor explains that having a right involves an entitlement to “…register complaints, demand redress, or call for legal enforcement of their rights…”[15] whenever they are violated. Because plants and non-human animals lack the capacity to carry out any of these entitlements, Taylor suggests that it would be nonsensical to recognize them as having rights.[16] Though each of these objections, if correct, would be conclusive on its own, taken together they seem devastating.

However, Taylor offers a way to understand rights that would allow us to preserve the notion that acting inconsistently with the attitude of respect for nature constitutes rights-violating behavior. Where the first view understood duties possessed by others as the corollary of rights recognized and asserted by the rights-holder, Taylor’s alternative conception of rights holds them to amount to recognitions of duties towards others which result from taking proper account of their inherent worth.[17] If failure to act in accordance with these duties indeed qualifies as rights-violating behavior, then actions inconsistent with Taylor’s attitude of respect for nature could certainly be understood in this way.

But do actions which fail to properly respect others’ inherent worth constitute potentially enforceable rights violations by themselves, or are the conditions described in Taylor’s first account really necessary for the recognition of rights? To answer this question, we might ask whether there are any seemingly clear cases where we would think ourselves justified in coercively enforcing the “rights” of a victim who does not fulfill one of the criteria mentioned in Taylor’s first view. And in fact, Taylor alludes to two groups of such entities: the insane and the severely mentally handicapped.[18]

Members of these two groups seem to fail on each of the four tests Taylor offers in his argument that plants and non-human animals do not have rights. Some insane and severely mentally handicapped individuals are certainly unable to conceive of “legitimacy,” to see themselves as being inherently worthy of respect, to choose whether to exercise or waive a right, to complain about violations of their rights, to demand restitution, and to call for the enforcement of their rights. Accordingly, Taylor’s first view would deny that we can possibly conceive of these individuals as having rights.

But nevertheless it seems clear that we would often think ourselves justified in using coercion to protect these individuals. And it does seem that the kind of situation in which we would feel justified in doing this lines up quite well with Taylor’s second conception of rights. That is, we coercively intervene to protect the insane and the severely mentally handicapped when we feel that they are being treated by others in a manner which fails to take proper account of their inherent worth. If we feel that this practice is based on a manner in which those individuals are entitled to be treated, then it seems to follow that we recognize them as having rights, in spite of all of the objections that Taylor’s first account has to offer, and that these rights are of the kind suggested by Taylor’s second account.

Of course, it could be that we are simply unjustified in behaving in this manner towards the severely mentally handicapped and the insane. But if we are justified, then it seems that we have good reason to think that coercion could be justified in protecting plants and non-human animals, whose inherent worth we have acknowledged by assuming that Taylor’s attitude of respect for nature is an objective account of the way that virtuous individuals will interact with non-human nature.

---

[1] Taylor, P.W., page 59.

[2] Ibid, pages 71-80.

[3] Ibid, page 89.

[4] Ibid, pages 172-192.

[5] Ibid, pages 88-89.

[6] Ibid, pages 15-16.

[7] Ibid, page 167.

[8] Ibid, pages 226-227.

[9] Ibid, pages 243-244.

[10] Ibid, pages 219-255.

[11] Ibid, page 246.

[12] Ibid, pages 246-247.

[13] Ibid, pages 247-248.

[14] Ibid, pages 248-250.

[15] Ibid, page 251.

[16] Ibid, pages 250-251.

[17] Ibid, pages 251-255.

[18] Ibid, page 225.