Many economists will find Yale psychology professor Paul Bloom’s new book very appealing. In Against Empathy, he frequently and favorably quotes Adam Smith, has nice things to say about economists, defends rationality, sees the importance of a utilitarian perspective, makes use of insights from public choice, and recognizes that the problem he sees with empathy is typically found in political settings. Yet I believe his case against empathy is seriously diminished in light of public choice considerations. Further, much of what he dislikes about empathy can be effectively dealt with by markets.

Empathy explained / Bloom spends much of his opening chapter distinguishing kindness, concern, compassion, love, and sympathy from empathy. These distinctions are appropriate; empathy—sharing in another person’s feelings—is not the same as other caring emotions. However, many of the problems that he blames primary on empathy also apply to other caring emotions, especially in the setting where he argues those problems are most pronounced—a setting that he calls “spotlighting,” which I discuss below.

Bloom points out that he agrees with Smith’s understanding of empathy, but emphasizes that Smith used the word “sympathy” to mean empathy as Blooom understands the word. Smith defines sympathy in The Theory of Moral Sentiments (1759) as the capacity to think about another person and

place ourselves in his situation … and become in some measure the same person with him, and thence form some idea of his sensations, and even feel something which, though weaker in degree, is not altogether unlike them.

That, says Bloom, is “how I am thinking about empathy.” In other words, if “your suffering makes me suffer, if I feel what you feel, that is empathy in the sense I’m interested in here.” This is often called “emotional empathy” and it “can occur automatically, even involuntarily.” This is to be distinguished from what is known as cognitive empathy, which is understanding that another is in pain without feeling the pain yourself. Unless specified otherwise, when Bloom uses the word “empathy” he is referring to emotional empathy.

Problems with empathy / Bloom accepts the perspective of most economists that “when it comes to morality … nobody can doubt that consequences matter.” While he recognizes that intentions are important, he focuses on what he sees as the consequentialist moral problems created by empathy. Among them is what he calls the “spotlight” effect: empathy focuses moral attention on

the one over the many. This perverse moral mathematics is part of the reason why governments and individuals care more about little girls stuck in wells than about events that affect millions or billions.

Consider three of Bloom’s examples of the spotlight effect. In 1949, the country’s attention was riveted on a 52-hour attempt to rescue 3‑year-old Kathy Fiscus after she fell down a well in San Marino, Calif. The country later mourned when the rescue effort failed. In 1987, the country’s anxiety focused on another child who had fallen down a well: 18-month-old Jessica McClure of Midland, Texas. Nationwide rejoicing followed her successful 55-hour rescue. Bloom’s third example comes from ethicist Peter Singer, who argues against donating to the Make-A-Wish Foundation, which finances short-lived “experiences of a lifetime” for very ill children. Singer points out that the money spent on one of these “dream days” would save more lives if given to the Against Malaria Foundation to be spent on mosquito-repellent bed nets in malaria-prone areas. Similarly, the money spent on the rescue attempts of little Kathy and Jessica could plausibly have saved more lives if spent in other, less dramatic ways.

In fairness to Bloom, he acknowledges that “empathy is not the only facet of our moral lives that has a spotlight nature. Emotions such as anger, guilt, shame, and gratitude are similar.” But this is his only mention of those emotions.

Instead of empathy, he counsels that people should act on compassion, as indicated by the book’s subtitle. He describes compassion as not

sharing the suffering of the other: rather, it is characterized by feelings of warmth, concern and care for the other, as well as a strong motivation to improve the other’s well-being. Compassion is feeling for and not feeling with the other.

The implication is that compassion is under our rational control to a greater degree than is empathy. He briefly acknowledges that when making some decisions, “many of my arguments against empathy apply to compassion as well.” Yet this comes after he tells us that when “struggling with a moral decision,” rather than relying on empathy when deciding “how to improve things,” it is “much better to use reason and cost-benefit analysis, drawing on a more distant compassion and kindness.” Leaving no doubt about how he feels about empathy, he tells us that “the capacity for emotional empathy, … defended by so many scholars, theologians, educators, and politicians, is actually morally corrosive.”

Critical to Bloom’s argument is his discussion of the connection of reason and rationality to morality. He mentions the late ethicist James Rachels, “who sees reason as an essential part of morality.” Bloom then considers the claim that “most people are incapable of rational deliberation.” In doing so he observes that those emphasizing irrationality suggest that “they themselves—and those they are writing for, you and me—are the exceptions” because they make rational cases for irrationality.

Bloom acknowledges human irrationality and accepts himself as being “only human” when stating that his book no doubt “contains weak arguments, cherry-picked data, sneaky rhetorical moves, and unfair representations of those I disagree with.” In other words, he is subject to confirmation bias (he doesn’t use that term) like everyone else. Yet he makes the convincing claim that (hard) “science provides an excellent example of a community that establishes conditions where rational argument is able to flourish.” I am somewhat less convinced with his follow-up statement that “the same holds, to various extents, in other domains, such as philosophy, the humanities, and even some sorts of political discourse.”

Emotions and political decisions / But can the spotlighting of an emotion—whether empathy, compassion, sympathy, etc.—on identifiable people be expanded to include a much broader set of people, so that help can be directed to where it does the most good?

Bloom is clearly aware of the difficulties in achieving such an expansion. He quotes Mother Teresa: “If I look at the mass, I will never act. If I look at the one, I will.” In his prologue he states that “we will never live in a world without empathy—or without anger, shame or hate, for that matter.” But he holds out hope that “we can create a culture where these emotions are put in their proper place, and this book is a step in that direction.” Yet he never indicates how that culture might emerge or be created, or what the proper place for the various emotions might be, other than that we should moderate the influence of empathy on political decisions.

In his closing chapter, “The Age of Reason,” he tries to make a hopeful argument that the political influence of empathy can be moderated. Bloom recognizes the difficulty of achieving such moderation, using some general arguments from public choice economics. These are arguments he admits “I am unhappy making,” possibly because he recognizes how they overwhelm his more hopeful suggestions on the potential for moral reasoning, as well as those by Steven Pinker, Robert Wright, Singer, and (yes) Smith, at least when applied to the political arena.

Bloom’s concern over moderating the political influence of empathy begins with his comment that “there are areas of life where we certainly seem stupid. Take politics.” His explanation for this “stupidity” is that unless we are members of a powerful interest group with a special interest in a political issue, our views on issues “don’t have to be grounded in truth, because the truth value doesn’t have any effect on [our lives].” He could have added that even when an issue does affect voters’ lives, human emotions will exert the dominant influence on how we vote because the probability of one vote decisively determining an election’s outcome is effectively zero.

It is difficult to see how he can maintain hope that political decisions can be influenced by fine distinctions between emotions. For example, a voter can realize personal satisfaction by voting for a policy she favors emotionally, while she can ignore with impunity any undesirable personal consequence of the policy because of the miniscule probability that her vote would determine whether it passes or not. Indeed, because the probability of a vote having a decisive effect on an election is effectively zero, the satisfaction from emotional expression is the primary driver of most voting decisions.

There can be little hope that a voter will make her decision only after determining whether her emotional satisfaction from voting a certain way is based on empathy, compassion, or some combination of those and other emotions, and then applying Bloom’s analysis. In the political process, differences in how particular emotions affect our decisions effectively disappear, which seriously diminishes Bloom’s case against empathy. The qualification that some political decisions can be decisively influenced with direct lobbying by organized interests does nothing to help his case because caring emotions such as empathy, compassion, and kindness have generally had little effect on decisions by organized interest groups to use their political influence to profit at public expense.

Considering markets / Many of the problems Bloom associates with empathy are addressed effectively by market information and incentives. Consider the following two examples of how markets overcome the spotlight effect and how they limit the negative effects of confirmation bias on rationality.

Markets are immune to the spotlight effect when appropriate to be so, although this is a virtue that often enrages many people. For example, those who work for or own stock in drug companies are no less sympathetic than the rest of us to the relatively few victims of rare diseases. However, market prices provide drug companies with the best information available on the tremendous human costs of developing drugs to benefit a few by diverting research away from developing drugs that benefit many. Politicians and journalists find advantage in spotlighting the heart-breaking stories of the identifiable few and excoriating the drug companies for ignoring or “overcharging” them. Yet far more human grief from illness is alleviated and more drugs to help the relatively few are developed in economies made more productive by the broad illumination that only market prices provide.

Next, consider the problem of confirmation bias. It is surely a less serious problem than behavioral economists and many psychologists would have us believe because of their own confirmation bias. No one would deny, however, that it is a serious problem in academics and politics, where feedback that could correct erroneous views is either absent or easily ignored. In markets, though, prices regularly make short work of even the strongest-held confirmation bias. For example, confirmation bias is seldom stronger than in the beliefs of entrepreneurs regarding the commercial value of their products. Yet when entrepreneurs confront market prices informing them that those products are worth less to consumers than they cost to produce, their confirmation biases quickly turn to dust and they begin to consider alternatives to their former convictions. If the confirmation biases of university professors and politicians were confronted with information as accurate and compelling as that communicated by markets, scholarship and public policies would improve as absurd ideas and policies are promptly discarded.

Conclusion / As an economist, I am sympathetic to logical arguments against popular policies and beliefs that are based almost entirely on their emotional appeal. Such beliefs as “Import restrictions increase domestic employment” and “Higher minimum wages reduce poverty” come to mind. But Bloom is arguing against a specific emotion (empathy) that he believes is particularly harmful because it misdirects benevolent efforts away from where they would do the most good.

There may be some truth to his argument that empathy is more likely to cause such misdirection with spotlight effects, and that it poses a greater threat to rationality and morality than other emotions such as compassion. Even so, given the exaggerated influence of all emotions in political decisions, particularly by voters, any additional destructive power of empathy over other caring emotions is hardly worth worrying about.

The problem is that all caring emotions, when exercised in the political process, are exercised without the relevant information and sense of responsibility necessary to avoid spotlighting, the irrationality of confirmation bias, and other unfortunate outcomes. In sharp contrast, when decisions can be made through markets (which admittedly is not always practical), spotlighting, confirmation bias, and irresponsible emotional influences are strikingly reduced and outcomes are more rational and broadly beneficial.

The most important conclusion to be taken from Bloom’s book is that the best hope for more rational and less emotional decisions is stronger constitutional limits—both procedural and substantive—on government. Those limits would make it more difficult for activities best left to markets to be shifted to the political process.