Just once in a while one would like to see them a little less disabused and bitter, with perhaps a dash of that naivete they are so bent on denouncing, with some openness to the unexpected, the possible.
Albert HirschmanFutility, perversity, jeopardy
In The Rhetoric of Reaction, published in 1991, Albert Hirschman sought to catalogue what he called “two hundred years of reactionary rhetoric” (p. 1). He identified three “reactionary” objections to reform proposals: perversity, futility and jeopardy (p. 7). In Hirschman's account, the three objections can be found in many times and places; they have a mechanical quality. He emphasized that perversity, futility and jeopardy theses do not come only from reactionaries. They can be invoked whenever a new proposal is offered. But he urged that “the arguments are most typical of conservative attacks on existing or proposed progressive policies” (pp. 6–7).
Of the three, the perversity thesis is the most dramatic (and clever); it suggests that “any purposive action to improve some feature of the political, social, or economic order only serves to exacerbate the condition one wishes to remedy” (p. 7). The power of the perversity thesis lies in the claim that the relevant reform “will produce, via a chain of unintended consequences, the exact contrary of the objective being proclaimed and pursued” – what Hirschman called “a daring intellectual maneuver” (p. 11). According to the perversity thesis, “the attack to push society in a certain direction will result in its moving all right, but in the opposite direction” (p. 11). If “everything backfires,” then (for example) “social welfare programs will create more, rather than less, poverty” (because people will not have an incentive to work), and efforts to require people to wear seatbelts will cost lives (because people will drive more dangerously) (pp. 12 and 41).Footnote 1 Hirschman shows that the perversity argument played a central role in reactionary responses to the French Revolution; in attacks on universal suffrage; and in objections to the modern welfare state, above all in the field of economics (pp. 12–35).
In drawing attention to the pervasiveness of the perversity claim, Hirschman did not mean to insist that the claim was always wrong. But he did insist that “it is unlikely to exist ‘out there’ to anything like the extent that it is claimed” (p. 35). On this count, Hirschman wrote bluntly: Could reactionaries “be embracing the perverse effect for the express purpose of feeling good about themselves? Are they not being unduly arrogant when they are portraying ordinary humans as groping in the dark, while in contrast they themselves are made to look so remarkably perspicacious?” (p. 36). He worried that “the effect is likely to be invoked for reasons that have little to do with its intrinsic truth value,” in part because “there are many unintended consequences that are welcome rather than the opposite” (p. 38). He insisted that the perversity thesis “has a certain elementary sophistication and paradoxical quality that carry conviction for those who are in search of instant insights and utter certainties” (p. 43).
The futility thesis is the simplest and perhaps the most intuitive; it suggests “that attempts at social transformation will be unavailing, that they will simply fail to ‘make a dent’” (p. 7). For those who invoke the futility thesis, any change will be “surface, façade, cosmetic, hence illusory, as the ‘deep’ structures of society remain wholly untouched” (p. 42). Hirschman illustrates the futility thesis with a joke from Eastern Europe after the installation of Communist government in the aftermath of World War II: “What is the difference between capitalism and socialism?” The answer: “In capitalism, man exploits man; in communism, it's the other way round” (p. 44).
Tocqueville's elaborate work on the French Revolution can be seen as a case study in the futility thesis, suggesting as it did that the magnitude of a seemingly epochal change was relatively modest (Hirschman, Reference Hirschman1991, pp. 45–50). Many attacks on universal suffrage have urged that it would be a cosmetic reform and achieve little or nothing (pp. 50–60). So too, reactionaries have objected that the real beneficiaries of the welfare state are not the poor, but “other social groups with more clout,” including “a large group of administrators, social workers, and sundry professionals … wholly dedicated to expanding their bureaus and perquisites” (pp. 60–69). In all cases, the objection is that “the basic structures of society” are unaltered (p. 72). Hirschman has something direct to say to those drawn to the futility thesis (he could have said it to those drawn to the perversity thesis and the jeopardy thesis as well): “Just once in a while one would like to see them a little less disabused and bitter, with perhaps a dash of that naivete they are so bent on denouncing, with some openness to the unexpected, the possible” (p. 80).
The jeopardy thesis is the least straightforward; it suggests that “the cost of the proposed change or reform is too high as it endangered some previous, precious accomplishment” (p. 7). The basic idea is that while the proposed change might be desirable in itself, it “involves unacceptable cost or consequences of one sort or another” (p. 81). The Principle of the Dangerous Wedge is an example; it is more familiarly known as a “slippery slope” argument, which means that if you do X (when X is admittedly desirable), you will also end up doing Y (when Y is truly terrible) (Hirschman, Reference Hirschman1991, p. 82 citing Cornford, Reference Cornford1908). On one view, for example, a small step toward regulating gun ownership should be resisted, because it will lead to the confiscation of guns. On another view, democracy threatens liberty (as, for example, when majorities want to punish dissenters). On another view, the welfare state threatens liberty (as, for example, when it results in violations of freedom of contract). On another view, modern regulatory agencies threaten democracy (because they give power to unelected bureaucrats). Or consider these words from Gustave Le Bon:
Had democracies possessed the power they wield to-day at the time of the invention of mechanical looms or of the introduction of steam-power and of railways, the realisation of these inventions would have been impossible, or would have been achieved at the cost of revolutions and repeated massacres. It is fortunate for the progress of civilisation that the power of crowds only began to exist when the great discoveries of science and industry had already been effected (Reference Le Bon2002, p. 35).
Friedrich Hayek often invoked the jeopardy thesis: “Freedom is critically threatened when the government is given exclusive power to provide certain services – power which, in order to achieve its purposes, it must use for the discretionary coercion of individuals” (Hirschman, Reference Hirschman1991, p. 114 quoting Hayek, Reference Hayek1939). Hirschman did not claim that the jeopardy thesis was always false, but he responded to it with a combination of bemusement and rage. Because the thesis is offered much more often than it turns out to be true, “there must be some inherent intellectual attraction in advancing” it (p. 164). Sometimes a reform does not, in fact, jeopardize some other reform. Sometimes it creates a kind of complementarity or mutual support. For example, the Earned Income Tax Credit might supplement other antipoverty programs and not compromise them at all. Graphic warnings on cigarette packages might supplement and even spur cigarette taxes, and not displace them. Democracy might turn out to promote innovation, rather than reducing it. A single reform, said to jeopardize some other practice or option, might actually make it more likely or more real, or larger still.
Hirschman almost titled his book “The Rhetoric of Intransigence,” and while that would have been an unlovely title, it was his central preoccupation, and his final chapter is called, simply, “Beyond Intransigence.” In that chapter, Hirschman takes off the gloves. He suggests that with his procedure in mind, “some ‘deep thinkers’ who had invariably presented their ideas as original and brilliant insights are made to look rather less impressive, and sometimes even comical” (p. 164). He notes that his “more basic intent” was “to establish some presumption, through the demonstration of repetition in basic argument, that the standard ‘reactionary’ reasoning, as here exhibited, is frequently faulty” (pp. 165–166). One reason is that the arguments “are invoked time and again almost routinely to cover a wide variety of real situations” (p. 166). Hirschman's ultimate goal “is to move public discourse beyond extreme, intransigent postures,” in “the hope that in the process our debates will become more ‘democracy friendly” (p. 168).
Hirschman's book appeared over 30 years ago, but it is directly relevant to countless current debates. In some ways, ours is the Era of Reactionary Rhetoric, in precisely Hirschman's sense. To reduce poverty, for example, a nation might significantly increase the minimum wage, but (it is said) the effect of the reduction would be to discourage employers from hiring people at the bottom of the economic ladder; the effect would be perverse. To reduce gun violence, a nation might require background checks for gun purchases, but (it is said) the requirement would do no good at all; it would be futile. To combat the risks associated with COVID-19, a nation might impose a mask mandate, but (it is said) that step would jeopardize individual freedom. Debates over greenhouse gas reductions are characterized by the futility thesis (if just one nation reduces emissions, it will do no good!) and the jeopardy thesis (if we limit greenhouse gas emissions, we will punish poor countries, and poor people, the most!). We can find the perversity, futility and jeopardy arguments in countless debates – and while they might come most frequently from the right, they are common on the left as well (as in the claim that nominally progressive actions, such as increases in funding for social welfare programs, will do little or nothing to combat structural poverty or structural racism).
From the standpoint of behavioral science, the rhetoric of reaction might usefully be explored in terms of motivated reasoning or more particularly “desirability bias” (see Tappin et al., Reference Tappin, van der Leer and McKay2018). The basic idea is that people are far more likely to believe things to be true if they want those things to be true. For some people, it is delightful to think that some reform will be futile or perverse, and it is upsetting or destabilizing to think that the same reform will be highly effective. A claim of futility or perversity is highly welcome. We can easily read Hirschman to have drawn attention to the motivated character of the rhetoric of reaction, and to have suggested that motivations should be put to one side in favor of investigation of actual evidence.
Nudges and nudging, very briefly
Nudges are interventions that steer people in particular directions but that also allow them to go their own way (see Thaler & Sunstein, Reference Thaler and Sunstein2021). A reminder is a nudge; so is a warning. A GPS nudges; a default rule nudges. Some nudges are educative; consider labels and warnings. Other nudges are architectural; consider automatic enrollment or website design that places certain options first or in large font. To qualify as a nudge, an intervention must not impose significant material incentives. A subsidy is not a nudge; a tax is not a nudge; a fine or a jail sentence is not a nudge. To count as such, a nudge must fully preserve freedom of choice. If an intervention imposes significant material costs on choosers, it might of course be justified, but it is not a nudge. Some nudges work because they inform people; other nudges work because they make certain choices easier; still other nudges work because of the power of inertia and procrastination.
There has been a great deal of work on the actual effects of nudges (see, e.g., Arno & Thomas, Reference Arno and Thomas2016; ; DellaVinga & Linos, Reference DellaVinga and Linos2022; Mertens et al., Reference Mertens, Herberz, Hahnel and Brosch2022). In some cases, they have a massive impact; changes in the default rule, from opt-in to opt-out, sometimes fall in that category (see Momsen & Stoerk, Reference Momsen and Stoerk2014, p. 380; Jachimowicz et al., Reference Jachimowicz, Duncan, Weber and Johnson2019, p. 164; Sarin, Reference Sarin2019, p. 1560; Bergman et al., forthcoming Reference Bergman, Lasky-Fink and Rogers2022). In some cases, nudges have a modest but real impact; disclosure of information (see Ellison et al., Reference Ellison, Lusk and Davis2013), reminders, and changes in “framing” sometimes have statistically significant effects, but they are hardly massive (see Choudhary et al., Reference Choudhary, Shunko, Nutessine and Koo2019, p. 16; DellaVinga & Linos, Reference DellaVinga and Linos2022, p. 92). In some cases, nudges have no measurable effect at all (Elbel et al., Reference Elbel, Kersh, Brescoll and Dixon2009; Kristal & Whillans, Reference Kristal and Whillans2020, p. 171; Dewies et al., Reference Dewies, Schop-Etman, Rohde and Denktaş2021, p. 221; Rela, Reference Rela2022). In some cases, nudges have “boomerang effects”; they produce the opposite of their intended result (see, e.g., Martineau et al., Reference Martineau, Ogilvie, Roland, Suhrcke and Kelly2011, p. 264; Bolton et al., Reference Bolton, Dimant and Schmidt2020, pp. 3–4; Osman et al., Reference Osman, McLachlan, Fenton, Neil, Löfstedt and Meder2020).Footnote 2 In some cases, nudges have the intended effect, but they also impose costs, because (for example) people suffer welfare losses as a result of the intervention (see Allcott & Kessler, Reference Allcott and Kessler2019; Thunstrom, Reference Thunstrom2019). – the “ruined popcorn” effect, as we might call it (see Sunstein, Reference Sunstein2019). The welfare measure is the right one (see Sunstein, forthcoming Reference Sunstein2023), and while we know less than we should about the welfare effects of the full universe of nudges, many of them almost certainly increase welfare.Footnote 3
At the present time, many of the various findings offer a positive picture of the effects of existing nudges (see Benartzi et al., Reference Benartzi, Beshears, Milkman, Sunstein, Thaler, Shankar, Tucker-Ray, Congdon and Galing2017; Allcott & Kessler, Reference Allcott and Kessler2019; Thunstrom, Reference Thunstrom2019; DellaVinga & Linos, Reference DellaVinga and Linos2022, p. 90), while also suggesting vindication, on some occasions, of the perversity thesis, the futility thesis and the jeopardy thesis. Some nudges do have little or no impact; some nudges do compromise other values; a few nudges do aggravate the very problem they are meant to solve. It is consistent with Hirschman's plea for “democracy friendly” debates to suggest that we need much more in the way of empirical testing. In fact we might offer a (friendly) amendment to Hirschman's own rhetoric. A great advantage of empirical testing is that we end up with answers and inquiries, not debates. We can identify testable hypotheses and then test them. We might ultimately end up with a kind of “map” to various effects, with an understanding of what we are likely to find, and when we are likely to find it (see Cadario & Chandon, Reference Cadario and Chandon2020, p. 459). If so, we might not “debate” at all, and our debates, if they occur, might well be relatively low-key. They would not involve intransigence. On the contrary, they would be characterized by openness and learning.
Of tweaks, slippery slopes and distractions
Some discussions of nudges and nudging are not low-key. In some cases, the perversity thesis, the futility thesis and the jeopardy thesis have been in something like screaming bright lights.
As applied to nudges, the futility thesis consists of the claim that nudges are mere “tweaks,” and have little or no effect on social problems (Ben-Shahar & Schneider, Reference Ben-Shahar and Schneider2014). On one view, they should be bracketed or abandoned for that reason (see Ben-Shahar & Schneider, Reference Ben-Shahar and Schneider2014; Nature Human Behavior Editorial, 2020, p. 121). If, for example, the goal is to improve financial choices, disclosure requirements might do little or nothing (see Ben-Shahar & Schneider, Reference Ben-Shahar and Schneider2014). If the goal is to increase savings in retirement, automatic enrollment in pension plans might do far too little; a savings mandate might be much better (see Bubb & Pildes, Reference Bubb and Pildes2014, p. 1609). If the goal is to reduce obesity, calorie labels might be thin gruel (so to speak); they might have no effect at all (see Kiszko et al., Reference Kiszko, Martinez, Abrams and Elbel2014, pp. 1252–1255). If the goal is to combat climate change (it might be said, and has been said), the focus should be on carbon taxes, subsidies, and regulatory mandates, not on home energy reports and energy-efficiency labels.Footnote 4
The jeopardy thesis takes several forms. On one view, the choice architects who try to come up with good nudges lack the necessary information; they face the same “knowledge problem” as do planners of all kinds (see Glaeser, Reference Glaeser2006, p. 144). For that reason, they are likely to blunder, in the sense that they will nudge people in the wrong directions (see Rebonato, Reference Rebonato2012). Nudges are often a form of paternalism, and likely to be ill-considered for that reason (Rizzo & Whitman, Reference Rizzo and Whitman2020). Individuals care about a large number of things, not only (for example) health and safety, and if outsiders nudge people to be healthier and safer, they will compromise what individuals actually care about.Footnote 5 On another view, nudges are or can be a form of manipulation, and in that sense, they might compromise autonomy. They might formally protect freedom of choice, but they might also operate as de facto mandates (Bubb & Pildes, Reference Bubb and Pildes2014, p. 1599).
The perversity thesis is the most distinctive. For example, Ryan Bubb and Richard Pildes have argued that automatic enrollment in 401(k) plans has often turned out to be perverse, in the sense that in the aggregate, it leads to a situation in which savings go down rather than up (Reference Bubb and Pildes2014, p. 1609). The reason is that even if automatic enrollment leads to higher participation rates, a low default contribution rate (3 percent) can lead people to save less than they would if they made choices on their own – which means that a significant increase in participation rates can lead to a significant reduction in total savings. Simply put: More nudging, less savings. Pointing to the risk of both futility and perversity, Bubb and Pildes argue in favor of mandates rather than nudges; in their view, mandates are more likely to be successful, and less likely to have boomerang effects.Footnote 6
A generalization of the perversity argument comes from Nick Chater and George Loewenstein, who claim that focusing on individual behavior is something like a catastrophic error (Reference Chater and Loewenstein2022, pp. 3–5). Chater and Loewenstein believe that it is important to distinguish between “i-frames,” focused on individuals, and “s-frames,” focused on “the system of rules, norms and institutions by which we live” (p. 3).Footnote 7 In their account, nudges are typically i-frame interventions (see p. 9). (Importantly, nudges often do not focus on individual behavior at all; consider the greenhouse gas inventory, which is designed to nudge companies toward reducing their emissions, or “scorecard diplomacy,” which is designed to nudge nations. But I bracket that point here.) Chater and Loewenstein urge that i-frame interventions “don't fundamentally change the rules of the game, but make often subtle adjustments that promise to help cognitively frail individuals play the game better” (p. 3). They deploy both the futility thesis and the perversity thesis, with an emphasis on the latter (pp. 10–26). In their view, the effects of i-frame interventions are “often modest,” with an impact that is “slight, at best,” which means that we have “disappointing results,” and hence a form of futility (p. 3)Footnote 8.
In a passage that could have come directly from The Rhetoric of Reaction, Chater and Loewenstein argue that i-frame interventions “may be counter-productive by drawing attention and support away from the kinds of systemic changes required to address most public policy problems” (p. 5).Footnote 9 In their view, “the entire enterprise of behaviorally-inspired public policy interventions is misdirected – and has been inadvertently aligning with the objectives of opponents of s-frame reform” (p. 6). As they put it, “promoting such interventions will either fail and/or weaken support for the system-level interventions that threaten [corporate] interests” (p. 6). A strong clue (they think) is that “across a wide range of policy domains, corporate campaigns are persistently backing i-frame approaches, presumably believing that they are failing and will continue to fail” (p. 6).Footnote 10
Thus “many behaviorally oriented academics with an interest in public policy, including ourselves, have inadvertently reinforced the ineffective i-frame perspective,” and “behavioral i-frame interventions have generally had disappointing results, but more importantly can have perverse unintended consequences, such as reducing support for policies solutions that are well known to be effective, and leading individuals to blame themselves for problems with systemic origins” (p. 27) (Here is the rhetoric of reaction in action, with the futility thesis followed “more importantly” by perversity thesis.) Chater and Loewenstein urge that behavioral scientists should shift to “s-frame” interventions by noting people's insensitivity to conflicts of interest; emphasizing the power of framing; and stressing people's power of adaptation (for example, to adapt to, and to accept, s-frame reforms that they might have resisted) (pp. 26–67).
In their central argument, Chater and Loewenstein offer a noteworthy variation on what Hirschman had in mind; call it an argument from distraction. The central idea, which can be applied to a host of reform proposals, is that if one focuses on ineffective or modest Reform A, one will divert time and attention from transformative Reform B, which means that one will end up, sadly and even tragically, with Reform A when one would have otherwise ended up with Reform B. The argument from distraction deserves an honored place in the rhetoric of reaction. Needless to say, it can be offered by the left or the right. If a reformer proposes a subsidy to promote purchases of electric vehicles, the proposal might crowd out discussion of a ban on gasoline-powered vehicles; if a reformer proposes cigarette warnings, the proposal might crowd out discussion of a substantial increase in cigarette taxes; if a reformer proposes automatic enrollment in retirement plans, the proposal might crowd out discussion of mandatory savings.Footnote 11
In principle, of course, the argument from distraction might be true, but it might also be false. It might even point in exactly the wrong direction. Support for Reform A might be a first step toward, and make possible, Reform B (a possibility to which I will return).
Testing
Hirschman was mostly concerned with rhetoric, not with substance. But the rhetoric of reaction raises substantive issues. What is the best way to deal with the three theses? The simplest answer is to test them empirically. What does the evidence actually tell us?
The futility thesis is the most straightforward. Do the relevant interventions have no effect? In fact? It is noteworthy that in discussions of many interventions, including nudges, the idea of futility does not generally refer to no effect. More often, it refers to effects that seem modest and perhaps disappointingly modest, especially in view of the sheer magnitude of the problem. For example, graphic warnings might prevent a very small number of the expected annual cancer deaths from smoking; fuel economy labels might do very little to address the problem of climate change; disclosure requirements might not do much to prevent credit card companies from imposing high late fees, overuse fees and interest rate charges; increasing take-up rates for the Earned Income Tax Credit might have no significant effect on the overall poverty rate. It is often possible to characterize an effect as modest, and an initiative as futile, if the benefit is very small compared to the magnitude of the underlying problem.
Beware of that characterization. Suppose, for example, that an antismoking initiative would prevent 1000 premature deaths annually. Such an initiative would have an exceedingly modest impact on the overall problem of premature deaths from smoking (estimated to be in the vicinity of 480,000 each year in the United States alone, and seven million globally), but an initiative is not exactly futile if it would prevent 1000 premature deaths annually. The monetized benefit of that savings (assuming a value of a statistical life of $10 million) is in the vicinity of $10 billion; producing $10 billion in benefits is not exactly futile. Or suppose that an initiative eliminated 100 million tons of carbon dioxide – a seemingly modest gain. If the social cost of carbon is set at $50, we are speaking of $5 billion in monetized benefits. A nudge that achieves $5 billion in monetized benefits is not futile. (Nothing I have said here is meant to deny that some nudges are futile. Some are; so are some subsidies and penalties.)
If a nudge ends up aggravating the very problem it is meant to solve, it is indeed perverse, and as noted, some nudges fall in that category (see Hall & Madsen, Reference Hall and Madsen2022). As we have seen, however, perversity in that sense is rare. To understand the perversity thesis, a key question is at what level of abstraction to describe the goal of the nudge.Footnote 12 If the goal is to improve social welfare, a nudge might be perverse even if it works to reduce caloric intake; people might be thinner but sadder. An increase in participation rates might seem to attest to successful nudging, but if participation makes people worse off, the effect will be perverse. As noted, the welfare question is central; we need to investigate it. That is true not only of nudges but also of subsidies, taxes, mandates and bans. Nudges designed to decrease alcohol consumption might be a good idea, on welfare grounds; a ban on alcohol consumption might be a bad idea, on welfare grounds.
What about i-frames and s-frames? What about the argument from distraction? If companies support Reform A and fight fiercely against Reform B, perhaps we can infer that Reform A is ineffective and that companies support it for that reason. But it is important to be careful here. Reform A might be good, all things considered, and Reform B might be bad, all things considered; companies might be right to support the first and to resist the second. (It is hazardous to think that a reform is good because companies dislike it, or that a reform is bad because companies like it, or think that it is in their self-interest to support it.) We cannot know without investigating the details. As Hirschman saw, the perversity effect might be a fairy tale. Many unintended consequences are welcome rather than the opposite, and reforms might be complementary or mutually supportive. We can imagine the following possibilities: (1) an i-frame intervention has no effect on the adoption of s-frame interventions; (2) an i-frame intervention makes it more likely that policymakers will make s-frame interventions; and (3) an i-frame intervention makes it less likely that policymakers will make s-frame interventions. Chater and Loewenstein are focused on (3), but they do not have anything like good evidence that it is a frequent occurrence (Reference Chater and Loewenstein2022).13 (Their admittedly interesting tales about the behavior of some interest groups – and speculations about issues of causation in the complex domain of policy outcomes – do not count as good evidence.)
We might go further. Many people think some version of (2): In their view, a nudge might be desirable in itself, but it will pave the way toward a mandate or a ban – a slippery slope argument. For some people, that is a problem; for others, it is something to applaud. Graphic warnings on cigarettes might, in fact, strengthen support for, or provide an opening for, increases in cigarette taxes; fuel economy labels might, in fact, strengthen support for, or pave the ground for, fuel economy requirements. As we have seen, some people emphasize a version of (3), pointing to the problem of distraction. To know whether that problem is real, we would like to have a randomized controlled trial, where the control condition involved a nudge (or i-frame) and then a set of outcomes, and the treatment condition involved no nudge (or i-frame) and then a set of outcomes. The question would be whether, in the treatment condition, we observe a set of outcomes that the nudge somehow discouraged or blocked – for example, an aggressive fuel economy rule, a carbon tax, a tax on sugar-sweetened beverages, a substantial increase in the Earned Income Tax Credit, a flat prohibition on cigarettes, a more progressive income tax, criminalization of sexual harassment, a ban on late fees and overuse fees.
Because history is run only once, there is no such randomized controlled trial. But it is worth pausing over the logic of the argument. Is it reasonable to suggest that if nudges were not in the picture, the United States, Canada and Italy would have more aggressive fuel economy rules? Would the Trump administration have favored such rules if fuel economy labels were not available? (In many nations, fuel economy nudges are complementary to aggressive fuel economy rules, and supported by exactly the same people; those who seek i-frame interventions also, and simultaneously, seek s-frame interventions. In the United States, far-reaching climate legislation, enacted in 2022, was not deterred by climate-related nudges, strongly favored by many of those who worked hard for that legislation.) Is it even plausible to suggest that if graphic warnings were not in the picture (so to speak), nations would ban cigarettes? Did New Zealand take steps toward banning cigarettes because it did not consider nudges instead? Is it plausible to suggest that if disclosure requirements were not on the table, nations would forbid credit card companies from imposing late fees and overuse fees? (The United States has heavily regulated such fees.) Is it reasonable to suggest that without fuel economy labels, the United States would have adopted a carbon tax?
The general point is that the number of factors that lie behind a decision to impose, or not to impose, a transformative reform of some kind (or what is not the same thing, to adopt an s-frame) is very large. It would be a rare case indeed in which the availability of a nudge (or an i-frame intervention) was either necessary or sufficient to block a more transformative (or s-frame) reform. It might be more plausible to suggest that the use of a nudge (or an i-frame intervention) helped spur support for a more transformative or (s-frame) reform. Recall Hirschman's worry that the rhetoric of reaction is often invoked when it has little “truth value,” and that “there are many unintended consequences that are welcome rather than the opposite” (Reference Hirschman1991, p. 38).
By-the-numbers
Hirschman was concerned with the by-the-numbers quality of reactionary rhetoric. As Hirschman had it, reactionaries are a little like the prisoners in the old joke:
A man goes to prison and the first night while he's laying in bed contemplating his situation, he hears someone yell out, “44!” Followed by laughter from the other prisoners.
He thought that was pretty odd, then he heard someone else yell out, “72!” Followed by even more laughter.
“What's going on?” he asked his cellmate.
Well, we've all heard every joke so many times, we've given them each a number to make it easier.
“Oh,” he says, “can I try?”
Sure, go ahead.
So, he yells out “102!” and the place goes nuts. People are whooping and laughing in a hysteria. He looks at his cellmate rolling on the ground with tears in his eyes from laughing so hard.
Wow, good joke huh?
Yeah! We ain't never heard that one before!
Hirschman wanted to demonstrate that all of us have heard the ones about perversity, futility and jeopardy many times before. Notwithstanding that fact, Hirschman insisted that reactionary rhetoric often produces a vigorous nod of the head – sometimes chastened, sometimes enthusiastic – even when it lacks the slightest empirical foundation. Hirschman's account also suggests that reactionary rhetoric can have an identifiable psychological source: an improbably triumphant combination of bitterness and despair.
From the right, reactionary rhetoric about nudges is marked by a cheerful insistence that all really is for the best in this best of all possible worlds. From the left, reactionary rhetoric about nudges offers something distinctive: an insistence, proud but evidence-free, that choice-preserving approaches are delaying the revolution. And these points are not, of course, limited to nudges; they can be found in objections, past and present, to reforms of many diverse kinds.
Acknowledgments
I am grateful to David Halpern and A.O. Oliver for terrific comments and to Lucia Reisch and Richard Thaler for valuable discussions. Special thanks to Julia Hammond and Heather Pincus for outstanding research assistance.
Competing interest
The author declares none.
