Steven Levitt and Stephen Dubner’s 2009 bestseller Freakonomics was built off of the idea that “incentives are the cornerstone of modern life” and that “economics, at root, is the study of incentives.” For Levitt and Dubner, and for the many scholars and readers inspired by the book, incentives are ubiquitous and come in all shapes and sizes, financial, social, and moral. It is impossible to understand the world and any choice situation without thinking in terms of incentives.

Ruth Grant’s new book is a welcome supplement, and in many respects a corrective, to this commonplace understanding of incentives. Grant agrees that incentives have come to be seen as ubiquitous, and here agrees with the Levitt and Dubner outlook. But she rejects the idea that incentives are in some way foundational. More pointedly, she does not view the idea of incentives to understand choice or human behavior as natural, or good, or benign. Strings Attached is an attempt to make visible the ethical dimensions of incentives and to provide a framework for distinguishing between legitimate and illegitimate uses of incentives.

The core idea is that incentives are best understood as an exercise of power, not as a characteristic of the choice situation between transacting parties in a voluntary exchange. The use of incentives is to be contrasted with other exercises of power, such as coercion and persuasion. We must evaluate incentives, Grant argues, by asking not whether a trade or action was genuinely voluntary but whether the exercise of power in deploying incentives was legitimate.

The book has eight brief chapters, all clearly and engagingly written. Grant roots the modern use of incentives in a historical account of the practices of scientific management of industry in the early twentieth century; incentives emerged as an effort at social engineering of the workplace (Chapter 2). She then provides her definition of an incentive: an intentional and deliberate offer by an external party (a person, organization, or political institution) designed to elicit a particular response with the aim of altering behavior (Chapter 3), and next develops a framework for distinguishing ethical from unethical uses of incentives (Chapter 4). She then moves to apply her understanding of incentives and her framework for evaluating them to a series of cases in which incentives are routine: plea bargaining in criminal justice, recruitment of subjects in medical research, loan conditionality by the International Monetary Fund, and motivation of children to learn (Chapters 5 and 6). The final chapters summarize and gesture at a new set of questions concerning the effect of incentives on democratic self-governance (Chapters 7 and 8).

Grant’s understanding of the use of incentives as an exercise of power is compelling. The book succeeds in undermining the more conventional and benign view of incentives as simple features of transactions between voluntary agents. It succeeds as a corrective to the Freakonomics view of the world. But what of the author’s arguments concerning how to distinguish ethical from unethical uses of incentives? The framework she defends for evaluating incentives has three criteria (pp. 50–51). First, understood as an exercise of power, does the incentive serve a legitimate purpose? Second, does it allow a voluntary response? Third, what are its effects on the character of the parties involved? Grant later upends this otherwise straightforward framework, calling these criteria necessary but not sufficient for an evaluation of the ethics of an incentive. We also must consider, she says, issues of multiple and competing purposes, effectiveness, impact on institutional culture, fairness, and undue influence (p. 74).

It is a complicated and, in some ways, unsatisfying framework. The first criterion—legitimacy of purpose—seems to beg the question. It asks us to evaluate the legitimacy of incentives by inquiring after the legitimacy of the purpose of the incentives. The second criterion—voluntariness—is important, but it is the same criterion at work in the economic understanding of incentives: Is an exchange between consenting parties genuinely voluntary? So this criterion does not distinguish Grant’s approach for understanding and evaluating incentives from the approach she is most concerned to resist. We are left with the third criterion—effect of character—as her most interesting and novel framework. The idea here is that the deployment of incentives can have a negative effect on the character of the individuals subject to the incentive. Incentives, like market dynamics more generally, can crowd out morality.

Consider, for instance, the tax incentives available in most countries to engage in charitable giving. Our ordinary understanding of charity is that it is a moral virtue, a sign of a generous character. But introduce a tax incentive into the mix and charitable giving can seem partially self-serving, an attempt to diminish one’s taxes in addition to being altruistic. A donor must ask him- or herself, am I giving because I am motivated by the tax deduction or by undiluted altruism? In short, Grant is right that incentives can sometimes have deleterious effects on character, and that these effects are obscured when incentives are viewed strictly as a market mechanism.

The payoff of Grant’s argument is in her exploration of four case studies. Reflecting the complexity of her framework, in none of them does she deliver an unambivalent verdict regarding the ethical use of incentives. Plea bargains are ethically suspect because they stand in tension with the purpose of criminal punishment, except when plea bargains yield, as a condition of their execution, important information about other criminals. Recruiting medical subjects into experimental trials with incentives is ethically acceptable in most cases but not all. IMF loan conditionality is permissible but worrisome because of undue influence and questions about the voluntariness of debtor nations in accepting loans. And incentives for children to learn are understandable but troubling because they potentially interfere with character formation, the creation of self-discipline, and the intrinsic motivation to learn.

Not only is Grant’s framework complex; the context for assessing each case is also complex, especially concerning measures of effectiveness of the incentive. Is the social science evidence clear, with causal, not mere correlational, effects of incentives shown on overall effectiveness? How should the time horizon for effectiveness be set? The upshot is that in each case, there is ample room for reasonable disagreement with the judgments reached by Grant, even on the terms of her own evaluative framework.

In the end, the major contribution of Strings Attached is found not in the particular judgments reached by the author about particular cases but in the compelling argument to resist the Freakonomics view of incentives as ubiquitous and simple features of choice situations. The deliberate deployment of incentives to alter human behavior is indeed an exercise of power, and, as such, they demand our ethical attention. Grant’s orients us correctly, even if her framework for evaluating the ethics of incentives is not as powerful or as useful as she intends.