The tragedy of the carrots: economics and politics in the choice of price instruments.

• Author: Galle, Brian

INTRODUCTION I. BACKGROUND A. Carrots and Sticks Defined B. Economics of a Pigouvian Tax II. CARROTS AND STICKS AS PRICE INSTRUMENTS III. CONTROLLING NEGATIVE EXTERNALITIES A. General Analysis 1. Revenues 2. Income and output effects 3. Distributive considerations 4. Repeated-game incentives 5. Repeated games and mitigation B. Examples IV. ENCOURAGING POSITIVE EXTERNALITIES A. General Analysis B. More Examples V. THE POLITICAL ECONOMY OF INCENTIVE DESIGN A. The Tragedy of the Carrots B. Carrots for Key Programs? C. Committing to the Carrot-Free Diet CONCLUSION INTRODUCTION

Humans sometimes need a bit of encouragement to do good for other people. In a well-known passage of the Old Testament, Jehovah castigates the Israelites for failing to pay a tithe to support His temple, and notes that He has cursed their nation for their collective failures. (1) Then, lightening up a bit, He also promises three blessings for those who tithe. (2) Modem secular leaders offer more worldly rewards, in the form of a federal income tax deduction, for those who support churches and other goods shared jointly with others. (3) In this Article, taking a lead from Malachi, I ask: What about the curse? Why not incentivize charitable giving with a penalty provision, instead of giving away public money? More generally, when should incentives take the form of punishments, rather than rewards?

A more familiar path to the same questions would be to suppose that we were all to agree that the U.S. government should take steps to slow global climate change. How should we do it? We might cap directly the amount of greenhouse gases emitted--for example, by issuing a limited number of permits to emit those gases and letting firms exchange them. (4) Or we might change the price of greenhouse gas components, such as through a "carbon tax." (5) And, though it may not be obvious at first glance, a very similar approach to the carbon tax would simply be to pay polluters to stop, much as the United States did during the brief life of its "cash for clunkers" program, and as it continues to do now through a variety of renewable energy tax credits, home rehabilitation tax credits, and so on. (6)

What policymakers have failed to consider closely is: Which of the price approaches is better? Taking money from polluters, or giving to those who clean things up? The stick, or the carrot? Surprisingly, the environmental literature has given only glancing attention to that question, and outside the carbon tax debate it seems hardly to have been considered in a systematic way at all] This Article fills that gap.

As many readers know, there is a vigorous debate in the environmental literature on the first set of choices I just mentioned: whether pollutants can best be controlled by regulating their quantity, or instead by affecting their price. (8) Within the subset of price options, the choice of carbon-tax design might call to mind prominent recent debates over the use of so-called "tax expenditures," which consist mostly of arguments over whether the tax system should be used to implement spending programs. (9)

These questions are interesting and important, but so far they have mostly obscured the fact that there is a third key set of decisions to be made about regulatory goals that, like global climate change policy, attempt to grapple with the costs one group of society imposes on another. Such costs (as, again, most readers likely know) are commonly called "externalities," and at least since A.C. Pigou economists have known that when a consumer does not pay the full cost of consuming a unit of a good, she is likely to purchase more than society optimally would want. (l0) One mechanism for correcting that inefficiency is to change the price the consumer pays for that next unit of a good--its "marginal" price--to reflect the total cost society bears from its consumption, (11) As it turns out, there are two ways of accomplishing that task: tax the good, or pay the consumer not to consume it. (12) In other words, we can make the externality producer worse off than under the status quo: a stick. Or we can make the producer better off: a carrot.

Choices between carrots and sticks are hardly unique to environmental regulation. Indeed, they pop up anywhere we might use a price mechanism for overcoming externalities. Since externalities are one of a handful of fundamental justifications for government regulation, the carrot/stick problem, I will argue, is pervasive throughout government action. (13) Another prominent recent example is health insurance. Individuals who seek medical care when they lack insurance (or other means to pay) create fiscal externalities for other users of the health system: either paying customers pay more, or healthproviders take home less, to cover the expense of those who can't afford care. (14) The Affordable Care Act implements both a carrot and a stick to deal with this externality, and it applies those two tools selectively to two different populations. The poorest households get a carrot: they receive a subsidy to buy their own health insurance. (15) Everyone else gets a stick: they must buy health insurance, or pay a penalty tax. (16)

By one measure, penalties and subsidies, or what I've been calling sticks and carrots, are largely indistinguishable. One way to put this point is that there is no marginal difference between taxing you a dollar if you do something and paying you a dollar if you don't. Smoke that cigarette? One dollar, please. Throw it away? Here's a buck. Either way, the marginal cost of choosing to smoke your next butt, rather than discarding it, is one dollar. This equivalence turns on the concept of opportunity costs, which I explain in more detail in Part I.B.

But carrots can also differ importantly from sticks in their other economic and even moral effects. (17) Relative to present policy, a carrot transfers wealth from taxpayers to its recipients, while sticks have the opposite effect. This transfer can change the preferences of the regulated party, fill or drain government coffers, suit or offend our preferences for punishment and just distributions, and change the incentives of parties who are planning for the next change in regulation. Many of the individual components of this analysis are familiar from other literatures, such as debates over the law of takings or the best way to compensate parties affected by changes in legal rules.

What is novel about my argument here is that it synthesizes these other literatures to reach a global assessment of the relative merits of carrots and sticks as policy tools. (18) Prior analyses, to the extent they've considered the question at all, have tended to dismiss carrots out of hand because of their supposed propensity to encourage some actors to do bad in order to be paid to stop. (19) As I argue here, there is actually considerably more nuance to the problem, with a number of factors that favor carrots in some circumstances. Still, at least when it comes to discouraging negative behavior, I agree we should often prefer sticks. Sticks reduce the wealth of those who make the rest of us miserable, which makes sense both in terms of the effect of income on preferences and also in terms of our sense of justice. Sticks also replace other costly forms of revenue, where carrots instead put extra burdens on the treasury.

Another contribution I make is to extend my analysis to the production of positive externalities. (20) Nearly all the existing discussion related to the choice of price instruments focuses on the classic case of pollution and close analogues. (21) But there is no reason the same analysis cannot be extended to the production of social goods, such as charity or innovation. Why do we reward donors instead of punishing the tightfisted? As I show, there is a surprisingly good case for using sticks to produce positive externalities, although it is not as clear-cut as with negative externalities.

I also add to the existing lore by analyzing the political tragedies that lead us to choose carrots over sticks. (22) Once we recognize the basic economic structure of sticks as transfers from a concentrated interest group to society at large, it becomes fairly obvious that politics will tend to favor carrots. Less obviously, many aspects of judge-made law, such as the doctrines of standing, unconstitutional conditions, and the dormant Commerce Clause, also contribute inadvertently to our hunger for carrots. Prior commentators have treated these political preferences for carrots as a reason to use carrots, especially in high-stakes legislation. I argue instead that acceding to demands for carrots is like capitulating to blackmail: it only leads to greater incentives for bad behavior in the future.

My analysis of the politics of carrots also raises a point of particular interest for those who study federalism. At this point there is a standard list of policy arguments either for assigning government responsibilities to the national government or devolving them to states and localities. (23) Federalism commentators have mostly devoted their efforts to rebalancing among these well-known points. (24) I offer here a new reason for preferring central government to local: local governments face excessive pressure to use carrots, and carrots are often inefficient. Thus, while the existing federalism literature recognizes that the presence of externalities that spill across borders might be a justification for federal action, my analysis implies that even externalities that do not cross borders might better be handled by the central government.

In short, our practice to date of neglecting the importance of the choice between carrots and sticks has led to some unfortunate policy decisions. Thinking about that carrot/stick choice systematically can teach us some useful lessons for institutional design. Thus, I also use the synthesis I develop to cast a new light on a wide...