The Unexpected Role of Tax Salience in State Competition for Businesses

Introduction

In 2012, Amazon agreed to invest $130 million in building two fulfillment centers and to create 1,500 jobs in New Jersey in exchange for the state relieving Amazon of its sales-tax-collection obligations.1 That New Jersey took action to lure Amazon into the state is unexceptional; states have long competed with each other over mobile firms by providing specific firms with targeted economic development incentives to encourage those firms to invest in the states.2 Until very recently, however, few, if any, states had provided such incentives in the form that New Jersey provided to Amazon,3 a form that this Article labels “customer-based” incentives. Instead, targeted economic development incentives have traditionally taken the form of such things as income tax credits and property tax abatements.4 For example, in 2013, New Jersey’s close neighbor Maryland provided Amazon with $43 million worth of tax credits in exchange for Amazon’s opening a one-million-square-foot distribution center and employing one thousand people in the state.5 Though “traditional incentives” like those provided by Maryland are unremarkable, customer-based incentives are anything but.

This Article introduces customer-based incentives to the academic literature by arguing for their remarkability and their place in states’ efforts to lure mobile firms and the jobs and infrastructure they bring. After providing an overview of how customer-based incentives function and the apparent oddity of their use, I argue that predictable consumer biases cause individuals to react to customer-based incentives in ways that make those incentives more effective at luring firms to the offering state and more beneficial to society as a whole than traditional incentives. This conclusion explains why the emergence of customer-based incentives is not so odd after all and, at a higher level, demonstrates that the form of a tax provision can affect its substantive consequences; policymakers can improve the efficiency and equity of tax provisions, particularly tax relief provisions, by incorporating the lessons of behavioral research into the design of those provisions.6

Customer-based incentives and traditional incentives represent two ways a state can achieve the substantive policy of encouraging a particular firm to invest in the state. Through traditional incentives, the state provides the firm with direct tax relief;7 through customer-based incentives, the state provides tax relief to the firm’s customers when they transact with the firm.8 The difference between customer-based incentives and traditional incentives may appear irrelevant when actors are economically rational; a firm can adjust its prices to capture as much of the tax relief provided through either form of incentive as it prefers.9 Further, when one relaxes the assumption of rational actors, there also seems to be a real danger to Amazon or any similarly situated firm that it would not be able to capture the tax relief from customer-based incentives; its customers might not tolerate the firm raising pretax prices to capture the relief. Research into consumers’ perceptions of fairness in pricing confirms the likelihood of this result; thus, the tax relief from customer-based incentives should be expected to stick with customers, at least to some degree.10 Thus, the emergence of customer-based incentives is somewhat mysterious; why break from the status quo of traditional incentives in favor of what appears to be a less firm-friendly form of incentive?

The effects of tax salience on consumer behavior provide an answer to this mystery. Tax salience refers to the level of awareness taxpayers have of a tax provision.11 Thus, when a consumer wants to spend $1,000 on a new laptop from Amazon but the additional $60 of sales tax stops her from doing so, that sales tax is salient to her. Research into tax salience demonstrates that many tax provisions—particularly sales taxes—may be “undersalient” to consumers; consumers ignore these taxes to some degree.12 Returning to that same consumer, when the sales tax is undersalient to her, she might completely ignore that $60 of sales tax and purchase the laptop anyway, even though she ultimately spends $1,060 and exceeds her preferred budget. Other tax provisions can be “hypersalient” to consumers, meaning that consumers overreact economically to the provisions.13 The laptop purchaser might perceive a hypersalient sales tax as the economic equivalent of $120 instead of $60, making her even less likely to purchase the laptop. However, if tax relief can be made hypersalient to consumers, then they will overreact to its economic value; $60 of sales tax relief might feel like $120 to the laptop purchaser, causing her great satisfaction when purchasing from a firm receiving customer-based incentives, such as Amazon.

Hypersalience of the tax relief from customer-based incentives may seem outlandish at first; however, consumers’ behavioral biases make it not only feasible but likely. For instance, research has demonstrated that people suffer from a behavior termed “tax-label aversion”—people value not paying something labeled a tax just for the mere fact that it is labeled a tax.14 Because customer-based incentives offer consumers tax relief, those incentives trigger consumers’ tax-label aversion. Therefore, a firm receiving customer-based incentives can promote sales-tax-free shopping and expect a bump in demand because people simply do not like taxes.15 Traditional incentives do not offer this advantage because consumers are unlikely to think they are being relieved of taxes, even if the firm lowers prices to pass along the tax relief from traditional incentives. This suggests that, when it can invoke tax-label aversion in its customers, a firm can get more of a benefit from customer-based incentives than traditional incentives, explaining why Amazon would request customer-based incentives.16

Any hypersalient tax relief provided through customer-based incentives should be expected to benefit not only firms but also states, further explaining the emergence of customer-based incentives. Customer-based incentives are likely to generate more benefits for society than traditional incentives, all else being equal. These societal benefits result primarily from two sources. First, because of tax-label aversion, consumers feel more satisfaction when shopping at a firm receiving customer-based incentives than they would otherwise.17 Second, because a firm receiving customer-based incentives is less likely to capture the tax relief offered than a firm receiving traditional incentives, customer-based incentives are more likely than traditional incentives to undo the harmful effects to society of the original taxes.18 Taxes reduce the amount of beneficial transactions in a society and thereby reduce overall social welfare. When customers retain the tax relief provided by the state, it is as though the original taxes were never imposed; society is restored to a pretax world. If the firm retains the tax relief, as can be expected to happen in the case of traditional incentives, society remains in the after-tax world and the harmful effects of taxation are not eliminated.19

States have additional reasons to potentially prefer customer-based incentives. A common concern regarding traditional incentives is the fear of the tax relief being used to finance activities outside of the state—that the money the state lays out will somehow “escape” the state by the firm’s actions.20 Because the tax relief provided by customer-based incentives is directly tied to in-state consumption, those incentives are not plagued by the escape issue.21 Also, customer-based incentives provide tax relief—even if only nominally—to anyone willing to shop from the recipient firm, which may make customer-based incentives more equitable in public opinion.22 However, certain customer-based incentives involve a form of selective nonenforcement of taxes by the state that may be perceived as particularly unfair and politically undesirable (though the states’ current experiences should soften this concern).23 Customer-based incentives may further increase social welfare if they are perceived as more equitable and less subject to abuse than traditional incentives, but, at a minimum, these additional aspects of customer-based incentives further demonstrate how such incentives can differ from traditional incentives.

In sum, the goals of this Article are twofold: first, to introduce customer-based incentives to the academic literature; and second, to use those incentives to demonstrate that the form of a tax provision can affect its substantive consequences. To these ends, the Article proceeds in four Parts. Part I details the different forms of targeted economic development incentives, providing an in-depth look at how customer-based incentives function and describing their recent emergence and potential for growth. Part II then questions the puzzling emergence of customer-based incentives, the benefits of which appear more difficult for a firm to capture than those arising from traditional incentives. Part III offers a solution to this puzzle: the salience of the tax relief from the two different forms of incentives affects the benefits they produce for firms in ways that make customer-based incentives more appealing than traditional incentives. Part IV then examines the potential benefits of customer-based incentives to states, providing a basis for why a state would prefer customer-based incentives to traditional incentives as a matter of policy.

I. The Different Forms of Targeted Economic Development Incentives

Targeted economic development incentives have long been a part of state government spending, as states compete for mobile firms and the jobs and infrastructure they bring.24 This Part provides a brief overview of such incentives before narrowing its focus to customer-based incentives. Traditionally, targeted economic development incentives have provided specific firms with direct tax relief through such things as income tax credits and property tax abatements, like Maryland provided to Amazon,25 or with subsidies through such things as state-sponsored job training programs and infrastructure construction.26 In exchange, the state offering the incentives often hopes to bring (or retain) jobs, investment in infrastructure, and economic growth to the state, along with the additional tax revenues that accompany such things.27

This Article focuses on targeted economic development incentives providing tax relief. The overarching ability of such incentives to influence mobile firms’ decisions is the topic of much debate,28 which I do not enter. The fact remains that most, if not all, states offer targeted economic development incentives. Though many commentators conclude that the incentives’ impact on firms’ decisions is relatively small, they tend to admit that states offer the incentives anyway due to the perceived high-stakes competition for jobs and other drivers of economic growth.29 Targeted economic development incentives providing tax relief can take a number of forms. Traditional incentives are relatively uncomplicated; through them, the state provides direct tax relief to the firm, and the firm is then free to take advantage of that tax relief however it prefers.30 In contrast, customer-based incentives do not provide direct tax relief to the firm but instead provide tax relief to the firm’s customers.

The remainder of this Part details customer-based incentives more closely by first examining how they function and then inspecting their recent emergence. Though customer-based incentives are not necessarily limited to the area of sales taxes, the discussion will focus on sales-tax-based, customer-based incentives given that these are the only type of customer-based incentives provided by states so far. For these purposes, there are no meaningful differences between how a state imposes its sales tax or between sales taxes and use taxes.31 Therefore, for ease of discussion, all sales taxes and use taxes will collectively be referred to as “sales taxes.”

A. How Customer-Based Incentives Work

Understanding how customer-based incentives operate provides the necessary background for considering how they differ from traditional incentives. There are three parties directly involved in the provision of a customer-based incentive: the state, the firm, and the firm’s customers.32 The state provides the customer-based incentive, the firm receives the customer-based incentive, and the firm’s customers are relieved of their obligation to pay taxes on purchases from the firm. This tax relief for the firm’s customers is the defining attribute of a customer-based incentive, and it can conceivably take two forms: direct relief of the taxes imposed on purchases from the firm, and indirect relief from taxes. Customer-based incentives providing the former will be referred to as “direct customer-based incentives,” and those providing the latter as “indirect customer-based incentives.” The primary benefits of customer-based incentives—direct or indirect—to the firm are straightforward. Beyond the administrative cost savings that come with not having to collect sales taxes,33 the firm receives the competitive advantage of being physically present in the state34 and its customers not having to pay sales taxes on their purchases from it.35 This competitive advantage may result in both more customers and more profits per customer if the firm also raises its pretax prices to capture some of the tax relief offered by the incentives.

Direct customer-based incentives are rare, if not nonexistent. Such incentives presumably would be provided through statutory or regulatory language exempting purchases from a certain firm from sales taxes.36 In this scenario, there would be no sales tax owed on transactions with the firm, so the firm would not collect any taxes. Though direct customer-based incentives are rare, direct relief of sales taxes is not a foreign concept; such relief can be found in the statutes of all of the forty-five states (and the District of Columbia) that impose sales taxes. For example, exemptions for or lower tax rates on purchases of groceries,37 purchases from nonprofit organizations,38 or purchases in designated enterprise zones39 all provide direct tax relief to consumers. Though the goals of these exemptions may not always be economic development, they encourage consumers to buy groceries instead of restaurant meals, to buy from nonprofit instead of for-profit organizations, and to shop in enterprise zones instead of outside the zones. Enterprise zones—designated areas in which a state lowers tax rates and business regulations in an effort to stimulate economic growth and employment40 —offer a good example of direct tax relief for economic development purposes that direct customer-based incentives might imitate in a more targeted way. Indeed, enterprise zones provide a possible starting point from which direct customer-based incentives might evolve should states desire to venture into offering such incentives.41

In contrast, a state provides indirect customer-based incentives by relieving the firm of its obligation to collect sales taxes. Thus, the state does not formally relieve the firm’s customers of their obligations to pay sales taxes (the taxes are still owed), but instead the firm never collects the taxes, effectively relieving the customers of the taxes. This effective tax relief arises because individual compliance with sales tax obligations—technically, use tax obligations—is dismally low when a vendor does not collect the taxes at the time of sale, and the states’ efforts to overcome this low level of compliance have been largely ineffective.42 Thus, in the current state of affairs, if a vendor does not collect sales tax from an individual, the sales tax will not be paid; relieving firms of their sales-tax-collection obligations results in sales tax relief for their customers.

Though indirect customer-based incentives are a new phenomenon, most consumers have enjoyed a form of indirect tax relief for some time. Under Supreme Court precedent, remote vendors—vendors without a physical presence in the taxing state, such as online and mail-order vendors—are beyond the states’ taxing authority.43 Remote vendors do not collect sales taxes from their customers because they are not obligated to.44 And because individual consumers do not self-report and pay their sales taxes on their purchases from remote vendors, the consumers receive indirect tax relief on those purchases.45 According to one study, consumers’ failure to self-report and pay sales taxes not collected by online vendors contributed to an estimated $11.4 billion in lost revenues among all states in 2012.46 Anyone who has made a purchase online on which sales tax was not collected and who did not subsequently pay the tax has experienced this unintentional form of indirect tax relief, albeit by breaking the law.

B. The Emergence of Indirect Customer-Based Incentives

Though the future of indirect sales tax relief for purchases from remote vendors is unclear,47 indirect customer-based incentives are emerging. This Section examines this emergence and argues that such incentives are poised to grow in usage as a practical matter. In recent years, a handful of states have provided Amazon with indirect customer-based incentives by agreeing to delay the enforcement of Amazon’s sales-tax-collection obligations in exchange for Amazon’s commitment to create and maintain a certain number of jobs in the states and to invest certain amounts in facilities in the states.48 The states do not appear to have provided customer-based incentives (direct or indirect) to any firm other than Amazon yet;49 however, their use is ripe for analysis given their recent emergence.

That Amazon is the first mover does not mean that it is the sole candidate for customer-based incentives; other firms would benefit from them as well, and states should be expected to offer them to other firms as the potential benefits of the incentives—discussed in the following parts—become clearer.50 Even so, widespread use of the incentives might not be expected for a number of practical reasons. First, sales-tax-based, customer-based incentives necessarily have a diminishing return as more are offered. The competitive advantage that arises from the incentives relies on the firm not collecting sales taxes even though its competitors do.51 The higher the percentage of competitors collecting sales taxes, the larger the competitive advantage will be. While customer-based incentives will likely never reach the scale of traditional incentives, they can be tailored to address this problem. For one thing, they can be offered on a temporary basis, allowing the state to shift the incentives among firms over time.52 This might increase competition for the incentives, improving the advantages states receive in return. It might also decrease the value of the customer-based incentives to the first firm to receive them if the customer loyalty it gained is diminished by the sales-tax-free shopping made available elsewhere.53 A better solution to the problem of diminishing returns may be to limit the customer-based incentives geographically, to allow for multiple firms to receive them simultaneously in one state. Thus, customer-based incentives might best be administered on a local level.

Additionally, the reason a firm is seeking out targeted economic development incentives may affect whether customer-based incentives are appealing. If the firm needs predictable assistance in order to invest in the state, then customer-based incentives will not be as effective in luring the firm as traditional incentives because customer-based incentives do not provide block grants of known tax relief—the firm must engage in commercial activity before it sees any benefit from customer-based incentives and the size of that benefit remains uncertain. However, if the firm is looking to increase its returns, customer-based incentives may be appealing as they are not as concrete as traditional incentives tend to be—the more commercial activity the firm engages in, the higher the benefit from customer-based incentives. There is the potential that the benefit for the firm from the customer-based incentives will be larger than the set amount provided by traditional incentives.54

Finally, customer-based incentives appear to have a limited universe of potential recipients—vendors responsible for collecting sales taxes. However, this does not have to be the case. One can conceive of a state offering people income tax credits for engaging in transactions with a particular firm; similar general incentives are made available by the federal government for purchasing energy-efficient goods.55 What is important to the customer-based incentive form is that tax relief is provided to someone that interacts with a firm as a result of that very interaction with the firm. The firm may then figure out how to capture some of that tax relief if it pleases. Creative policymakers should be able to find ways to create customer-based incentives in a variety of contexts.

Thus, many of the practical limitations on the growth of customer-based incentives are not insurmountable. Customer-based incentives will likely never fully replace traditional incentives, but there is no reason that a state could not provide both to a firm. Policymakers should thus invest time in better understanding how the incentives work and ways to make them more effective.

II. The Puzzle of Customer-Based Incentives

Though customer-based incentives appear primed to take on a larger role in states’ economic development efforts, whether states should actually expand their use of such incentives may—or, at least, should—depend on whether the difference in form between customer-based incentives and traditional incentives has substantive effects. Otherwise, the simple administrative costs of adopting a new form of incentive should cause states to prefer to continue relying on traditional incentives. Before the next Parts analyze the potential benefits of customer-based incentives over traditional incentives, this Part argues that at least one difference between the two types of incentives makes the emergence of customer-based incentives appear quite odd. Because of predictable consumer biases regarding fairness and losses, firms should be expected to have more difficulty capturing the tax relief offered through customer-based incentives than that offered through traditional incentives.

The fact that customer-based incentives provide tax relief to a firm’s customers rather than directly to the firm does not necessarily mean that the firm will not receive any of the economic value of that tax relief. Because the taxes relieved are still levied generally, market prices should continue to reflect after-tax prices if the marketplace is competitive. If the firm controls its prices after receiving the incentives, it can keep them at the after-tax, pre-incentives level, capturing the full benefit of the tax relief.56 The ability of the firm to control prices depends on its relative competitive position to consumers.57 One of the key characteristics of targeted economic development incentives is that they provide firms with a competitive advantage by lowering their costs of doing business or the costs of doing business with them; therefore, it is assumed that the firms have strong competitive positions vis-à-vis their customers, as those firms are the only ones able to offer lower after-tax prices.58 Thus, firms receiving targeted economic development incentives may be expected to control the post-incentives prices of what they sell.

If it is assumed that the firm’s ability to control prices is entirely unimpeded, then there is no reason to think that the firm’s ability to capture tax relief provided through customer-based incentives is meaningfully different than its ability to capture tax relief provided through traditional incentives. Numerous theories have demonstrated that, when rational actors and no transaction costs are involved, the initial distribution of rights (or benefits, or burdens, etc.) among parties should not matter; the parties will transact with each other to reach their preferred balance of rights.59 It is not surprising that the form of targeted economic development incentives should not matter under the same assumptions; the firm will be able to adjust pricing to reach its preferred balance. However, relaxing the assumption that a firm has entirely unimpeded control over prices after receiving targeted economic development incentives introduces some stickiness to the pre-incentives prices. In other words, when the firm faces transaction costs to changing its prices, pre-incentives prices should be expected to remain in effect until the value of the tax relief captured by changing prices outweighs the transaction costs. Two major transaction costs to firms in adjusting their prices are considered here: the costs imposed by the fairness concerns of consumers and the costs imposed by the pain consumers feel from losing tax relief to which they feel entitled.60

A. The Fairness of Pricing Changes

Research demonstrates that people value being treated fairly and are willing to suffer economic harm to themselves in order to punish those they perceive to be acting unfairly.61 Perhaps the most well-known research in this area involves the ultimatum game experiment.62 In this experiment, two participants are given the opportunity to share a sum of money; one participant proposes a distribution between the two, and the other participant may either accept or reject the proposal.63 If the proposal is accepted, both participants get their proposed share of the money; if the proposal is rejected, neither participant receives anything.64 Contrary to the expectations of rational actor theory, participants routinely reject low-ball offers—they are willing to suffer economic harm to avoid being treated unfairly and to punish the first participant for making an unfair proposal.65 More complex iterations of the ultimatum game reach similar results.66 This suggests that if a firm is perceived as unfairly adjusting (or not adjusting) its prices after receiving targeted economic development incentives, customers may punish the firm by refusing to shop with it.67 A firm should therefore consider how consumers can be expected to react to price changes in response to the receipt of incentives and the potential costs of establishing unfair prices.68

Researchers have studied how consumers judge the fairness of price changes. In one influential study, Professors Daniel Kahneman, Jack Knetsch, and Richard Thaler conducted a series of telephone surveys in which they posed a number of scenarios to consumers to determine how the consumers judged the fairness of different pricing practices.69 This research demonstrated that consumers tend to judge the fairness of pricing changes based on the interaction of two reference points: the firm’s reference profit and the customer’s reference price.70 These reference points are established prior to the proposed change;71 for purposes of this analysis, it is assumed that the pre-incentives profit and price are the appropriate reference points (though as developed below, what exactly the pre-incentives price is in consumers’ minds is not clear). Further, the research demonstrates that customers feel entitled to their reference prices but permit firms to protect their reference profits at the expense of the reference prices;72 thus, it is perceived as fair for a firm to increase prices as the result of increased costs of doing business, but it is perceived as unfair for a firm to increase prices due to increased market power.73 Interestingly, people judge it fair for a firm to maintain prices even if its costs of doing business decrease.74 Thus, people expect to be charged the reference price, and fairness requires only that the firm not take advantage of increased market power to raise prices.

What do these findings mean in the context of targeted economic development incentives? Certainly, consumers are not going to object on fairness grounds to a firm lowering its prices. But will the firm face pushback for maintaining after-tax, pre-incentives prices? Traditional incentives provide direct tax relief to the firm, lowering its costs of doing business; thus, consumers’ fairness concerns should not affect the ability of a firm receiving such incentives to maintain pre-incentives prices and thereby increase its profits. For example, suppose a jeans-selling firm receives a traditional incentive. Prior to receiving the incentive, it sold jeans to Amy for a price of $100 and generated a profit of $20. The firm’s reference profit is $20, and Amy’s reference price is $100. If the traditional incentive provided the equivalent of $10 of tax relief for each sale, the firm would be able to maintain the $100 reference price and increase its profit to $30 on each sale without offending the fairness sensibilities of its customers—it would not be pressured to pass along a portion of the tax relief to its customers. Assuming a 5 percent sales tax, Amy would pay $105 total for her jeans regardless of whether the firm receives the traditional incentive. One might expect the $105 after-tax price to be Amy’s reference price, but whether Amy’s reference price is $100 (the pretax price) or $105 is irrelevant; when the firm maintains its pretax price, the after-tax price is the same as well.

On the other hand, maintaining after-tax prices after receiving customer-based incentives may trigger fairness concerns in consumers. This result will occur if consumers view the pretax price instead of the after-tax price as the reference price. Because customer-based incentives eliminate the sales taxes consumers pay, maintaining the pretax price does not result in the same after-tax price; to maintain the after-tax price and capture the tax relief, the firm would need to increase the pretax price it charges. As the firm faces no threat to its reference profit as a result of the customer-based incentives, consumers will be unwilling to sacrifice their reference price; the increase in the pretax price will be rejected on fairness grounds.

When consumers do not pay attention to sales taxes, the consumers should be expected to think of the pretax price as the reference price, as that is the price the consumers are basing their decisions on.75 Research into “tax salience”—which has garnered increasing amounts of academic attention in recent years76 —demonstrates that consumers often fail to fully take sales taxes into account before making purchasing decisions.77 Tax salience refers to the level of awareness a taxpayer has of a tax provision.78 This concept can be split into two components: first, awareness of the economic effects of the tax provision—“market salience”—and second, awareness that the tax provision is the source of those economic effects—“practical salience.”79 Though market salience, which the next Part returns to in further detail, and practical salience, which the next Section examines in more detail, are related and may inform one another, they are distinct concepts.80 To illustrate the difference between them, consider the example of liquor excise taxes which are included in the posted price of the liquor rather than separately stated. Professors Raj Chetty, Adam Looney, and Kory Kroft found that consumers responded economically to such taxes as one would rationally expect them to respond—an increase in tax lead to a proportionate decrease in demand.81 Thus, the excise taxes were fully market salient to the consumers. However, the researchers also found that the liquor consumers were not aware that the excise taxes were the source of price increases, blaming the business for the increases instead of the taxing state.82 This result indicates that the taxes were not practically salient to the consumers.

The tax salience research indicates that a firm receiving traditional incentives will not face fairness transaction costs when maintaining its pretax prices to capture the tax relief provided, but that a firm receiving customer-based incentives (at least sales-tax-based ones) may if it must raise pretax prices to capture the tax relief provided. Unless consumers can be convinced to adopt after-tax prices as their reference prices, the firm is less likely to capture the tax relief from customer-based incentives than that from traditional incentives. However, even if sales taxes become salient to consumers such that after-tax prices become their reference price, another bias—loss aversion—may hinder the firm’s ability to capture tax relief from customer-based incentives.

B. Practical Salience and Loss Aversion

As noted, practical salience refers to the awareness that a tax provision is the source of economic effects that a person experiences.83 If a tax provision is practically salient to a person, the person should be expected to react to the provision in such ways as protecting or attacking it both politically and in their interactions with nonpolitical actors.84 Exactly how tax provisions become practically salient and how people will react to practically salient taxes are often difficult things to predict.85 People are motivated to take action for a number of reasons, of which the practical salience of tax provisions is only one.86 Even so, this Section discusses different reactions that can be expected to arise from the practical salience of tax relief from the different forms of targeted economic development incentives, arguing that the lower practical salience of tax relief from indirect customer-based incentives than that from direct customer-based incentives explains why firms would prefer tax-collection-obligation relief rather than direct tax relief for their customers.

Studies indicate that factors affecting the practical salience of a tax provision include the burden the provision places on the taxpayer, particularly in terms of compliance;87 the transparency of how the provision works;88 and whether the provision directly falls on the taxpayer.89 Thus, the market salience of a tax provision is likely to inform its practical salience; if a taxpayer realizes the economic burden of the tax, the tax provision is more conspicuous than if the burden were not noticed. However, market salience does not fully define practical salience; even tax provisions that have a high degree of market salience may have a low degree of practical salience.90 Consider the excise taxes on liquor discussed earlier; the excise taxes were market salient but were not practically salient.91 Liquor consumers dissatisfied with the high price of liquor might not understand that a source of their dissatisfaction is the excise taxes, and therefore the consumers might not react to the taxes practically.

Tax relief from the different forms of targeted economic development incentives should be expected to have different levels of practical salience for consumers. Some basic level of practical salience of tax relief provided by targeted economic development incentives should exist for consumers; watchdog groups and government agencies issue reports regarding such incentives that spur people to take action for or against them.92 However, tax relief from traditional incentives should be less practically salient to consumers than relief from customer-based incentives because the tax relief from traditional incentives does not directly fall to consumers. Further, indirect customer-based incentives should be less practically salient to consumers than direct customer-based incentives because, although the tax relief from indirect customer-based incentives is directly experienced by the consumers, the process by which consumers receive that relief is not as transparent as the process behind tax relief provided through direct customer-based incentives.93 Thus, on a scale, traditional incentives will have the least amount of practical salience to consumers, indirect customer-based incentives will fall in the middle, and direct customer-based incentives will have the highest amount of practical salience, all else being equal.94

The lower practical salience to consumers of traditional incentives and indirect customer-based incentives may cause consumers to fail to perceive the full extent to which those incentives are the source of tax relief that flows through to them. On the other hand, the higher practical salience of direct customer-based incentives to consumers should cause consumers to be more aware that the tax relief they receive comes from those incentives. The practical salience levels of the various forms of targeted economic incentives are important because of a behavior known as loss aversion—reacting more strongly to losses than gains from the status quo.95

Research has demonstrated that people commonly display loss aversion, iterations of which are referred to as the endowment effect or the status quo bias.96 In one of the most famous experiments, researchers gave mugs to some participants and nothing to others.97 They then asked the participants how much money they would be willing to accept to part with the mugs and how much money they would be willing to pay to acquire a mug, respectively. The values placed on the mugs by those who were given them were significantly higher than the prices others were willing to pay to acquire them. When the roles were switched, the same people who earlier had placed higher values on the mugs they possessed indicated that they would not pay as much to acquire a mug, demonstrating that people place a higher value on losing things already possessed than gaining the same things. In the tax context, Professors Edward McCaffery and Jonathan Baron have demonstrated that people suffer from a form of loss aversion they term “penalty aversion”—people prefer things labeled bonuses, such as a child bonus, which come across as gains, to things labeled penalties, such as a childless penalty, which come across as losses, even if the end results are the same.98

When consumers view the status quo as the tax relief from targeted economic development incentives falling to them, loss aversion will inhibit the firm’s ability to capture that tax relief; the firm’s price increases would decrease demand for its goods more than expected as customers feel the pain of losing the tax relief they thought was theirs. The additional practical salience to consumers of the tax relief provided through direct customer-based incentives makes consumers more likely to view their receiving that tax relief as the status quo. On the other hand, because traditional incentives and indirect customer-based incentives frame the relief provided as that of the firm (either direct tax relief or the relief of a collection obligation), the tax relief consumers receive through those incentives appears as a gain to consumers instead of a loss.99 The status quo in these instances is the firm receiving an incentive and the customer paying sales tax. Further, statutory incidence can go a long way in defining the status quo;100 though a firm could conceivably promote its nonobligation to collect sales taxes when it receives either direct or indirect customer-based incentives, granting that collection relief to the firm on the books makes people more likely to view the status quo as the firm receiving a benefit.

Thus, because indirect customer-based incentives inhibit the impact of loss aversion, firms should be expected to prefer them over direct customer-based incentives. However, the mystery of why a firm would prefer customer-based incentives over traditional incentives in the first place still remains; unlike customer-based incentives, traditional incentives should trigger neither loss aversion nor fairness concerns in consumers. The following Part provides an explanation for this mystery by arguing that customer-based incentives provide a larger benefit to firms than traditional incentives, even if the firm is unable to capture the actual tax relief provided through the incentives.

III. The Tax Salience Solution to the Customer-Based Incentives Puzzle

State policymakers can implement their substantive policy goals in a variety of forms. For instance, a state wishing to encourage economic development by convincing a firm to invest in the state can provide that firm with traditional incentives or customer-based incentives (or some mix of the two). Without passing judgment on the effectiveness of targeted economic development incentives generally, this Part uses such incentives as a case study to demonstrate that the choice of a tax provision’s form is significant; different forms of taxation intending to implement the same policy can generate different substantive effects because they have different levels of salience to consumers. In particular, differing levels of market salience—the awareness of a tax provision that affects a person’s economic decision-making101 —should cause consumers to react more strongly to the tax relief provided through customer-based incentives than that from traditional incentives. Coupled with the preference for indirect customer-based incentives detailed in the prior Part, this result provides a solution to the puzzling emergence of indirect customer-based incentives.102

A. The Degrees of Market Salience

As alluded to, recent research has demonstrated that tax provisions have differing levels of market salience to taxpayers.103 Tax provisions can be undersalient, fully salient, or hypersalient.104 The idea of full salience serves as a baseline for measuring when a tax provision is undersalient or hypersalient—tax provisions that are fully salient to a taxpayer have rationally expected effects on the taxpayer’s economic decision-making. In contrast, tax provisions that are undersalient have smaller-than-expected effects on the economic actions of those affected,105 and tax provisions that are hypersalient have greater-than-expected effects.106

Theory suggests that a behavior termed “spotlighting” affects the market salience of a particular tax provision to the taxpayer.107 Spotlighting refers to the tendency of a person to focus on particularly conspicuous components of a price or transaction to the exclusion of other components, thereby misperceiving the total cost of the transaction.108 Thus, taxes paid manually by the taxpayer are more salient than taxes paid automatically,109 immediate taxes are more salient than delayed taxes,110 and aggregated taxes are more salient than broken-up taxes,111 all because the total cost of the tax becomes more conspicuous.

Sales taxes are a prime example of taxes that may be undersalient; consumers spotlight on tax-free posted prices and ignore the cost of the taxes.112 For example, in one of the more well-known studies, Professors Chetty, Looney, and Kroft ran a three-week-long experiment in grocery stores in which they included the sales tax in the posted price of goods on the shelves.113 This action caused a decrease in sales of the goods, indicating that people had not been paying full attention to the taxes before they were included but instead had spotlighted on the tax-free posted prices. The act of spotlighting on nontax elements of a price or a transaction alone does not necessarily mean that the tax will be undersalient to the parties involved.114 After all, a consumer that spotlights on pretax prices could be aware of the tax generally (that is, it could be practically salient to the consumer) but overestimate its burden. For example, Chetty, Looney, and Kroft posit that the market salience effects they observed in their grocery store experiment derived from consumers believing that calculating the burden of the taxes would be too costly, so they instead ignored the taxes and focused on posted prices.115 This theory requires the consumer to know about the tax but to underreact to it.

At least two predictable behaviors contribute to the likelihood that an ignored tax will become undersalient. First, people display “anchoring bias” when making estimates of unknown values.116 Anchoring causes a person’s estimates to gravitate toward a conspicuous number, even if that number has nothing to do with what is being estimated.117 For instance, in one study, people were asked to spin a wheel numbered 0 to 100 and then estimate various percentages, such as the percentage of African countries in the United Nations.118 Despite the clear arbitrariness of the numbers on the wheel, people’s estimates gravitated toward the number they spun.119 When consumers spotlight on pretax prices, those prices will set an anchor for their estimates of the tax they will owe, causing the consumers to underestimate the taxes.120

Additionally, people display “optimism bias,” which causes them to underestimate the likelihood of losses and overestimate that of gains.121 For example, optimism bias may lead consumers to disregard consumer product safety warnings because the consumers undervalue the possible harms from the product.122 An optimistic consumer that has not calculated the actual value of the taxes that she will owe may similarly underestimate that value.123 However, most of the research on optimism bias focuses on relative risks—the risk an individual faces when compared to other people—and taxes may not fit this mold, potentially weakening the expected effect of optimism bias in the context of the market salience of tax provisions.124 Even so, if a consumer is considering the probability that a certain tax rate will apply, she may be inclined to expect a lower tax rate due to optimism bias.

It should be noted that the research on the salience of sales taxes has focused primarily on shopping in physical stores, not over the Internet.125 As noted, Chetty, Looney, and Kroft theorized that calculating the burden of the taxes would be too costly for consumers, leading them to ignore the taxes and focus on posted prices.126 Education is thought to be one of the primary methods of making taxes more salient,127 and the Internet offers vendors the ability to make sales taxes more salient by offering quick information on after-tax prices, reducing consumers’ calculation costs and making the taxes more immediate.128 Therefore, sales taxes may be more salient on Internet purchases, though the taxes can remain broken up and not necessarily immediately available (they may not show up until the customer goes to a checkout page having already made the decision to buy), so some undersalience concerns may still exist.

Finally, some tax provisions may be hypersalient, affecting taxpayers’ economic actions to a greater degree than expected from rational actors.129 The deduction for charitable contributions has been described as hypersalient because it encourages taxpayers who are not eligible to claim the deduction to make charitable contributions; taxpayers spotlight on the deduction, ignoring the limitations on claiming it.130 As a result, people make more charitable contributions than might be rationally expected from an economic point of view.

This market salience research indicates two ways that the forms of targeted economic development incentives can affect firms differently. First, as discussed above, when consumers adopt pretax prices as reference prices because sales taxes are undersalient to them, firms will have difficulty capturing tax relief from customer-based incentives because of consumers’ fairness concerns.131 Second, if either form of targeted economic development incentives is undersalient or hypersalient to consumers or to firms, then their responses to those incentives will not meet economically rational expectations. The following sections explain why tax relief from customer-based incentives should be hypersalient to consumers but tax relief from traditional incentives should not, causing customer-based incentives to generate more demand for the firm’s goods.

B. The Market Salience of Tax Relief from Targeted Economic Development Incentives to Consumers

Because firms control prices, they also control the amount of tax relief that consumers receive;132 it is the market salience of this amount of tax relief to consumers that is important. If the firm passes no tax relief onto consumers, then there is nothing for the consumers to be aware of and react to economically. The tax relief from the different forms of targeted economic development incentives that is passed on to consumers can be expected to have different levels of market salience to consumers; relief from traditional incentives should be fully salient, but relief from customer-based incentives should have some degree of hypersalience. This difference in salience is what causes customer-based incentives to generate more demand for the firm’s goods than traditional incentives.

The tax relief from traditional incentives should be fully market salient to consumers because when a firm lowers its prices to pass along the tax relief, the economic effect of the relief is immediately available and visible to consumers. There is no need to self-compute the value of the tax relief, nor is the relief disaggregated or delayed. In short, the value of the tax relief is not economically hidden in any way from consumers (they may not realize that the tax relief is the source of the price decreases, but that misperception relates to practical salience, not market salience). Recall again the excise taxes on liquor studied by Chetty, Looney, and Kroft: these taxes were not separately stated to consumers, but instead were included in the posted price of the liquor. This inclusion made the taxes fully market salient—consumers’ preferences responded to increases in the taxes as expected.133

On the other hand, the tax relief provided by customer-based incentives could realistically be undersalient, fully salient, or hypersalient to consumers. However, because consumers are predisposed to finding tax relief hypersalient, as discussed below, and it is in both the state’s and the firm’s interest for consumers to find that tax relief hypersalient,134 hypersalience is the expected result.

Consumers might find the tax relief from customer-based incentives undersalient through the same mechanisms that they find certain sales taxes undersalient. When consumers spotlight on pretax prices and make their purchasing decisions before being exposed to taxes, they underestimate the value of those taxes.135 Like those taxes, the tax relief from customer-based incentives is not immediate and could be overlooked by a consumer spotlighting on pretax prices. Though the consumer may be pleasantly surprised to find that she received tax relief for taxes she was not paying attention to, that tax relief would not affect her economic decision-making.

Similarly, the tax relief provided by customer-based incentives might be fully salient to consumers in the same way that other sales taxes are fully salient.136 As noted, Chetty, Looney, and Kroft posit that spotlighting on pretax prices occurs because consumers think that the personal cost of calculating after-tax prices is too high for the benefit obtained.137 However, consumers could be easily debiased by providing them with information on after-tax prices prior to their decision-making process through such mechanisms as posting after-tax prices in stores or displaying a tax calculator or after-tax prices online.138 Providing consumers with immediate information on post-incentive prices or the explicit amount of tax relief provided to them should make that tax relief fully salient.

Finally, consumers could find the tax relief provided by customer-based incentives hypersalient—they might overreact economically to the tax relief. This result may seem fanciful initially, but consumers are primed toward this result because of two behaviors: tax-label aversion and optimism bias. Research demonstrates that people are affected by tax aversion—they have negative preferences for taxes.139 A subcategory of tax aversion is tax-label aversion, in which people value the relief of charges labeled taxes; the tax label itself affects people’s preferences.140 For example, Professors Abigail Sussman and Christopher Olivola, through a series of surveys, found that people were (i) more likely to drive thirty minutes out of their way to buy a television sales-tax-free than to drive the same distance for a slightly larger price discount, (ii) willing to stand in line longer for a tax-free purchase than an identically discounted purchase with tax, and (iii) more likely to prefer tax-free municipal bonds than other bonds, even when the final economic return was equivalent.141 Other commentators have observed overconsumption resulting from sales tax holidays offered by states.142 The satisfaction people feel from avoiding taxes adds to their desire to transact with a firm receiving customer-based incentives, causing what appears to be an overreaction to the actual value of the tax relief provided by customer-based incentives. All the firm needs to do to enjoy the resulting increased demand is make the consumers aware that they can shop sales-tax-free if they shop with the firm.143 In this vein, many online vendors have promoted sales-tax-free Internet shopping as a major reason to purchase from them instead of local vendors.144

In addition, if the consumer understands that she will be receiving sales tax relief, but does not precisely know the amount of the relief, the consumer’s optimism bias may cause her to overestimate the value of the tax relief. As noted, this bias causes people to underestimate losses and to overestimate gains.145 Thus, for any given reference price, consumers can be expected to underestimate the losses from having to pay sales taxes and to overestimate the gains from the tax relief provided if they understand that such taxes and tax relief will be in place but do not know the value of those things. The same actions a firm takes to trigger its customers’ tax-label aversion could also trigger this overestimation of the relief if the firm avoids providing any hard numbers regarding the amount of tax relieved.146 Further, assuming the customer is focusing on the idea of receiving tax relief, anchoring bias may not affect the consumer’s estimation of the tax relief because there is no initial value given to the consumer on which to anchor.147 However, a consumer might anchor to posted prices if the idea of tax relief is not conspicuous enough; this result might lessen the hypersalience of the tax relief.

Though the actual scale of the hypersalience of tax relief from customer-based incentives is uncertain, because research indicates that consumers can—and are likely to—find such tax relief hypersalient, customer-based incentives can be expected to generate more benefit for the firm than traditional incentives.148 In fact, the firm may benefit without altering its prices at all (though the option to do so remains); it can maintain reference prices and still be more competitive than its peers.149 The increased demand for the firm’s goods resulting from any hypersalience of customer-based incentives’ tax relief thus should lead the firm to prefer customer-based incentives over traditional incentives of the same amount.150 For a firm like a remote vendor, already experiencing the effects of such hypersalience, customer-based incentives offer the opportunity to maintain that status quo after investing in the state;151 traditional incentives do not.

C. The Market Salience of Tax Relief from Targeted Economic Development Incentives to Firms

Might the market salience of tax relief from targeted economic development incentives to firms affect which form is preferable? This Section argues that, because firms’ economic reactions to the two forms of targeted economic development incentives are likely to be similar, those reactions should not lead to the forms generating different effects. When discussing the firm’s economic reactions to the various incentives, it is more accurate to consider the reactions of the managers, owners, and customers of the firm, as the firm itself does not have awareness.152 The reactions of the customers were discussed in the prior Section.

The market salience of the tax relief provided through traditional incentives is unlikely to be misperceived by the managers of the recipient firm because the relief is negotiated, direct, and immediate.153 The negotiation process offers the managers the opportunity to overcome any existing misconception of the firm’s tax burden by requiring them to consider the value of the tax relief being offered (and thus the initial tax imposed), as well as any limitations on it. Further, the managers’ rationally expected response to the tax relief is buoyed by the fact that the relief is provided directly to the firm and is immediate in the sense that the managers are aware of it prior to making any business location decisions and pricing decisions. Similar to the case with traditional incentives, tax relief from customer-based incentives should be fully salient to the firm’s managers. Though the tax relief is not provided directly to the firm, its value is not obscured to the managers by that fact because the firm must negotiate for the incentives and the incentives are also immediate.

In the same way that consumers should be expected to find the tax relief from traditional incentives fully market salient—because the relief is passed along to them through changes to posted prices154 —a firm’s owners should find the tax relief from either form of targeted economic development incentives to be fully market salient. Neither form of incentives causes tax relief to fall directly to the owners; relief is passed through to them only in the form of dividends or changes to the value of their ownership interests. These objective measures of value offer little opportunity for owners to misperceive the economic value of the tax relief provided to their firms. Thus, no differences arise in how either form economically affects firms as a result of their behavior because managers and owners of firms should be expected to react similarly to both forms of incentives.155 Therefore, firms’ preference for customer-based incentives should remain intact.

IV. Customer-Based Incentives as Sound Policy?

Firms may prefer customer-based incentives, but what advantages might customer-based incentives offer states?156 This Part argues that customer-based incentives have the ability to generate more benefits for society than traditional incentives because consumers feel more satisfaction from the tax relief provided by customer-based incentives and because customer-based incentives are more likely to reduce the negative effects of the taxes they relieve. Further, there are strong reasons to believe that customer-based incentives will be seen as more equitable than traditional incentives because of the way they provide tax relief, strengthening the case for states to prefer customer-based incentives. To these ends, this Part addresses three significant issues that the form of the incentives might affect: the social welfare each form might generate, the “escape” of tax relief from the state’s economy, and the erosion of civic virtue that results from the nonenforcement of tax laws.157

A. The Effects of Targeted Economic Development Incentives on Social Welfare

An issue for states to consider when deciding what form of targeted economic development incentive to offer a firm is the potential effects of the different forms on social welfare. Social welfare refers to the overall well-being of society—the accumulation of the benefits generated by interactions within society.158 Tax provisions typically diminish the amount of social welfare by generating what is referred to as “deadweight loss”—the social welfare lost because the cost of taxes causes people to stop engaging in otherwise-beneficial transactions.159 Figure 1 presents a basic graphical representation of these concepts: the “vendor surplus” represents the benefits the vendor receives from selling a good; the “consumer surplus” represents the satisfaction the customer receives from purchasing the good; in combination, the vendor surplus and consumer surplus represent the total amount of social welfare generated; the “deadweight loss” triangle represents the social welfare lost because of the additional cost of taxes imposed on the sale of the good, and the “tax revenue” rectangle represents the amounts of vendor surplus and consumer surplus that the state claims through the taxes. Tax provisions can also affect the distribution of social welfare among the members of society either progressively (redistributing from the better-off to the worse-off), regressively (redistributing from the worse-off to the better-off), or proportionally (not redistributing at all).160

Figure 1.  Social Welfare with Tax Provisions

In general, policymakers should prefer policies that maximize social welfare and that distribute that welfare in an equitable manner,161 and this Section is devoted to showing that traditional incentives and customer-based incentives can generate different effects on social welfare solely as a result of their different forms. This is a critical insight because it shows that form matters when implementing substantive tax policy; a state can improve social welfare by carefully designing a tax provision. In order to focus solely on the expected effects on social welfare of the form of the incentives, this Section relies on two key assumptions. First, it is assumed that the different forms of targeted economic development incentives are equally effective at achieving the goal of boosting the state’s economy by creating jobs, encouraging investments in infrastructure, and enlarging economic activity.162 That is to say, a dollar of tax relief provided through a traditional incentive will result in the same amount of economic growth as a dollar of tax relief provided through a customer-based incentive. A corollary to this assumption is that the two forms of incentives will bring the same amount of economic detriment to the recipient’s competitors.163 Second, it is assumed that a state providing targeted economic development incentives will respond in the same way to its reduced tax revenues regardless of the form which those incentives take. In other words, to finance the incentives, the state will cut spending or raise taxes in the same way for traditional incentives and customer-based incentives. Under these assumptions, the only remaining relevant consideration is that traditional incentives deliver tax relief directly to the firm and customer-based incentives deliver tax relief to the firm’s customers.

1. Targeted economic development incentives and social welfare.

Can targeted economic development incentives increase or decrease the deadweight loss associated with the taxes they relieve?164 Can they affect the distribution of social welfare? If the answer to either question is “yes,” then policymakers should be careful when designing their tax provisions in order to most effectively achieve their goals. The answers to the above questions depend on a number of factors. The first factor is whether one of the parties is able to control who ultimately receives the tax relief provided by controlling post-incentive prices.165 As discussed earlier in the Article, it is assumed that a vendor receiving targeted economic development incentives will have this control.166 If the vendor controls prices, it could keep them at the pre-incentives level, capturing the full benefit of the tax relief.167

The second important factor is the elasticity of the supply and demand curves—how much supply and demand change in response to price changes. The more elastic a supply or demand curve is, the more extreme responses are; a small increase in price would significantly reduce demand and significantly increase supply for a highly elastic good.168 Elasticity matters in determining who bears the burden of a tax; the party with the more elastic response to price changes will bear less of the tax burden.169 Thus, if demand for a good is more elastic than supply, consumers bear less of the tax burden than vendors, as demonstrated graphically in Figure 2.

Figure 2.  Social Welfare with More Elastic Demand

Considering these two factors together, it becomes clear that targeted economic development incentives can—but will not always—affect social welfare. To reach this conclusion, it is important to recognize that tax relief provided through targeted economic incentives does not necessarily result in the elimination of the deadweight loss created by the taxes. Because the taxes are still levied generally, market prices should still reflect after-tax prices. Therefore, the firm can keep prices at after-tax prices, and the deadweight loss will remain.170 Because of this result, it may be helpful to think of tax relief as a refund of taxes rather than the elimination of taxes; for the taxes to be refunded, they must have been imposed in the first place.

Targeted economic development incentives will have no effect on the amount of social welfare if it is in the firm’s interest to keep prices at the after-tax, pre-incentives level (price P2 in the Figures); such incentives will increase social welfare if it is in the firm’s interest to reduce prices to the pretax level (price P1 in the Figures), sharing the relief with consumers and eliminating the deadweight loss from the tax.171 To see when these two results should be expected, a comparison of Figure 1 to Figure 2 is helpful. In either case, a firm receiving incentives will recapture the vendor surplus that had gone to the state as tax revenue;172 the important consideration is how much consumer surplus is available in comparison to the additional vendor surplus the firm could generate by eliminating the deadweight loss. Keeping prices at the pre-incentives level P2 allows the firm to capture the consumer surplus that had previously gone to the state as tax revenue; lowering prices to the pretax equilibrium level P1 allows the firm to generate additional social welfare (and thus vendor surplus) by eliminating deadweight loss, though the firm will not capture any consumer surplus. In Figure 1, the consumer surplus going to the state is greater than the vendor surplus eliminated by the deadweight loss (the lower portion of the deadweight loss triangle); thus the firm should prefer to keep the pre-incentives price and capture the consumer surplus that had gone to the state as tax revenue. The deadweight loss will remain. However, Figure 2 demonstrates that as the elasticity of demand increases, the amount of vendor surplus eliminated by the deadweight loss grows and the amount of consumer surplus going to the state decreases. At some point, the amount of eliminated vendor surplus will exceed the amount of consumer surplus captured by the state, and the firm will prefer to return to the pretax price P1, eliminating the deadweight loss and increasing the amount of social welfare.

Targeted economic development incentives can also affect the distribution of social welfare. When a firm captures all of the tax relief, there is a redistribution of welfare from consumers to the firm (that is, the firm captures the consumer surplus that had previously gone to the state as tax revenue). Whether this distribution is progressive, proportional, or regressive will depend on the status of the consumers and how the welfare is distributed by the firm. As a firm is not an individual, the social welfare it captures will ultimately fall to others, such as owners of the firm (through increased valuation of the firm), workers at the firm (through increased compensation), or customers of the firm (through lower prices).173 If the firm does not capture the tax relief, but instead eliminates the deadweight loss from the original tax, the distribution of social welfare returns to the pretax balance. Again, the progressivity, proportionality, or regressivity of this distribution will depend on the status of the consumers and how the vendor surplus is distributed by the firm.

Because the discussion has so far assumed that the firm and consumers are economically rational actors with full awareness of the incentives provided, there is nothing to indicate that customer-based incentives and traditional incentives affect social welfare differently due to their different forms. This result is unsurprising. Recall that numerous theories have demonstrated that the initial distribution of rights among parties should not matter in a rational-actor model; the parties will transact with each other to reach their preferred balance of rights.174 The form of targeted economic development incentives should not matter under the same assumptions. However, relaxing these assumptions exposes how the different forms can be expected to have different effects on social welfare.

2. Customer-based incentives’ ability to increase social welfare.

This Section argues that customer-based incentives are more likely than traditional incentives to increase social welfare, both by eliminating deadweight loss and by increasing consumer surplus. These results arise from the restrictions consumer biases place on firms’ ability to control prices and potentially from the hypersalience to consumers of the tax relief customer-based incentives offer.

As discussed, maintaining after-tax prices after receiving customer-based incentives may trigger fairness concerns or loss aversion in consumers.175 These results inhibit a firm from altering its pricing so as to capture the tax relief customer-based incentives offer.176 Such concerns and the resulting inhibitions are not present in the case of traditional incentives.177 Thus, the tax relief from customer-based incentives is more likely than that from traditional incentives to be shared with the firm’s customers; external factors pressure the firm into doing so. As the prior Section detailed, when consumers retain the tax relief, it is as though the taxes were never imposed—deadweight loss is eliminated, and social welfare is generated. Therefore, the state should have some preference for customer-based incentives over traditional incentives.

Additionally, that consumers can, and are likely to, find the tax relief provided by customer-based incentives hypersalient may cause customer-based incentives to generate more social welfare than traditional incentives.178 When the tax relief is hypersalient, consumers derive more satisfaction from purchasing from the firm than from other vendors, so the demand curve shifts up.179 This shift in demand (which does not occur in the case of the fully market-salient tax relief from traditional incentives) generates additional consumer surplus, and thus additional social welfare.180 In other words, a smaller amount of customer-based incentives can achieve the same result as a larger amount of traditional incentives because customer-based incentives cost the state less. Figure 3 demonstrates this effect by building on Figure 2, which shows a situation in which the firm would elect to return prices to the pretax equilibrium point. However, the potential for increased social welfare as a result of the hypersalience of the tax relief from customer-based incentives relies heavily on the assumption that traditional incentives and customer-based incentives would burden the recipient’s competitors in the same amount.181 If customer-based incentives result in more harm to the recipient’s competitors than traditional incentives (as might happen if customer-based incentives affect the relative demand for the recipient’s products more than traditional incentives), the net effect on social welfare of customer-based incentives may be less positive.

Figure 3.  Social Welfare with Hypersalient Customer-Based Incentives

Even so, this Section has presented a number of reasons to suspect that customer-based incentives can generate more social welfare than traditional incentives. Thus, the use of customer-based incentives should not be dismissed out-of-hand on social-welfare grounds. So long as customer-based incentives have a better upside than traditional incentives in this regard, customer-based incentives should be appealing to states seeking to improve net social welfare. However, the overall effects on social welfare of any targeted economic development incentive will ultimately depend on a large number of factors of which the form of the incentive is only one. For instance, though this analysis has assumed such factors away,182 providing customer-based incentives to certain concentrated industries may prove particularly harmful to overall social welfare, as increased demand for one firm permits it to push its competitors out of business. A state seeking to improve net social welfare must balance the impact of the choice of firm with the impact of other such factors.

The following sections take a closer look at two such factors—the “escape issue” and the “civic virtue issue”. These are examined because they are commonly raised issues regarding the fairness of traditional incentives. People may derive more satisfaction, thus generating more social welfare, if the form of incentive selected better addresses these issues than the other does. In other words, people may be more satisfied when the form of tax provision selected is viewed as more equitable than other forms. The following discussion further demonstrates that the two forms can have different substantive effects and suggests that there are strong reasons to suspect that customer-based incentives will be considered more equitable than traditional incentives, improving their appeal.

B. The Escape Issue

Some critics of traditional incentives argue that the design of traditional incentives allows the firm to use the tax relief provided to finance activities in other states or to leave the state before fulfilling its obligations, causing the benefits of the incentives to “escape” the state in a way that is unfair to the state’s residents.183 In a sense, the firm using the tax relief to finance outside activities is merely a bargaining issue; the state should have required the firm to do more in the state for the incentives if it expected more. Money is fungible, and the fact that a firm uses tax relief to finance activities elsewhere does not mean that the firm did not fund its activities in the state. However, if a firm leaves the state before fulfilling its obligations, then the escape problem becomes more legitimate. In any event, a state’s citizens may not be comfortable with the possibility that any state-provided tax relief escapes the state.

As a basic response to the escape issue, many states have added various clawback provisions to their traditional incentives offerings which require the firm to return some portion of the incentives if it fails to meet its obligations.184 However, clawback provisions are rife with issues and have proven somewhat ineffective.185 As a primary matter, including the provisions decreases the value of the tax relief to the firm due to the risk that the firm may have to pay back some of the relief in the future, so more tax relief may have to be provided.186 Additionally, legal disputes can arise over whether the provisions apply when a firm leaves the state, creating costs for both parties and decreasing the likelihood that the state would be made whole even if it were entitled to be.187 Further, states may be reluctant to include or enforce clawback provisions because of the message doing so would send to other firms it hopes to lure in the future. One of the justifications for offering targeted economic development incentives is to show how business-friendly the state is; cutting against that display may be harmful to the state’s economic development goals.188

Customer-based incentives are less likely than traditional incentives to suffer from the escape issue, and thus do not need the support of awkward clawback provisions. The tax relief provided through customer-based incentives is directly tied to in-state activities, and to the extent that the relief is passed through to customers (as the above analysis indicates is likely), the relief is less likely to escape the state because individual consumers are likely to be less mobile than firms. To the extent the tax relief is captured by the firm, there may be larger escape concerns; however, because their tax relief is tied to in-state consumption, customer-based incentives discourage the firm from ceasing to engage in economic activity in the state. Finally, because the tax relief from customer-based incentives is spread out over time (due to the nature of sales and use taxes), the firm receives no lump-sum relief as it might through traditional incentives, mitigating the possibility of a misalignment in favor of the firm between what the state has paid out and the investments the firm has made in the state. If anything, the state is paying later for investments now when it provides customer-based incentives, so there is little risk of those payments financing activities outside of the state. Therefore, customer-based incentives appear far more capable of addressing the escape issue than traditional incentives and may be preferred by members of the public for that reason.

C. Civic Virtue: The Nonenforcement Issue

The final issue addressed here is the impact of the different forms of targeted economic development incentives on civic virtue. Civic virtue in this context refers to the willingness of people to comply with the law. Research demonstrates that such willingness is fostered by the consistent administration of laws that the public generally views as in line with their values.189 People express civic virtue when they feel that the law treats them fairly and equally with respect to everyone else.190 When civic virtue is eroded, people lose faith in the government and may begin to evade the laws.191 These types of actions are harmful if the laws are generally welfare enhancing.

Targeted economic development incentives have a built-in feature that potentially makes them damaging to civic virtue—they provide benefits to one person at the expense of others. Many may view this as unfair, and opponents of incentives often point out the inequity of providing a competitive advantage to one firm.192 Can the form of targeted tax incentives be expected to mitigate or exacerbate this corrosive effect on civic virtue?

Certain aspects of customer-based incentives indicate that the answer to this question is “yes.” Customer-based incentives may be less corrosive of civic virtue because the tax relief they offer is theoretically available to the entire public. Each member of the public can control whether or not she receives that tax relief (whether the firm then captures it is another question), which may make the application of the incentives appear more equitable.193 This fact may cause people to be more accepting of customer-based incentives rather than traditional incentives.

However, there is a meaningful difference between how traditional incentives and direct customer-based incentives provide tax relief and how indirect customer-based incentives provide tax relief. Traditional incentives and direct customer-based incentives provide tax relief by nontaxation—the law specifically provides relief of the tax at issue. For instance, a law providing a direct customer-based incentive would state that sales tax does not apply to sales by the firm. In contrast, indirect customer-based incentives provide tax relief by nonenforcement of existing taxes—by law, the tax is owed, so the state must be complicit in its failure to enforce the law.194 This complicity in the nonenforcement of existing laws disrupts the consistent application of the law, and thus indirect customer-based incentives may be more harmful to civic virtue than traditional incentives and direct customer-based incentives.195

Thus, the effects of customer-based incentives, and particularly of indirect customer-based incentives, appear more volatile than those of traditional incentives, which should affect civic virtue only to the extent people think that targeted economic development incentives are unfair. However, concerns about the effects of targeted economic development incentives on civic virtue may not be that serious in practice. Though traditional incentives are commonly criticized on equity grounds for favoring one firm over others, they do not appear to have significantly eroded civic virtue to the point that they cause taxpayers to evade states’ laws.196 As direct customer-based incentives similarly involve nontaxation, fears of their potential to erode civic virtue should be tempered. Additionally, states have engaged in the nonenforcement of sales taxes on Internet purchases for some time,197 with no apparent significant erosion of civic virtue outside of that specific scenario.198 Therefore, nonenforcement resulting from indirect customer-based incentives may also fail to significantly erode civic virtue generally. However, further research is required on this issue. States have typically demonstrated some desire to collect sales taxes on Internet purchases, at least outwardly showing a commitment to enforcement.199 This display may temper the effects of nonenforcement on civic virtue; the complicity involved in indirect customer-based incentives may be a more powerful eroding force.200

D. The Policy Appeal of Customer-Based Incentives

Though any number of additional factors influencing the effect on social welfare of targeted economic development incentives could be considered, this Part has argued that there are a number of reasons to support customer-based incentives as a policy matter. States appear locked into competition for mobile firms and the benefits they bring.201 Traditionally, the states have relied on direct spending in the form of traditional incentives to lure these firms into their states.202 However, customer-based incentives appear to have the potential to be more effective at attracting these firms while at the same time better supporting the state’s welfare, and thus should be taken seriously as a policy option.

In a way, customer-based incentives can be thought of as not being about tax relief for a firm at all; rather, they are about creating a motivated customer base for the firm in order to encourage it to invest in the state. Many argue that targeted traditional incentives are ineffective at influencing mobile firms’ business location decisions because so many other factors, such as the presence of natural resources, an educated workforce, and good infrastructure, are more important.203 Just as natural resources are important to a manufacturer, a motivated customer base may be important to a firm; customer-based incentives provide consumers with that motivation. By understanding customer-based incentives better, states may be able to better direct their economic development efforts. At a minimum, state policymakers must pay attention to the effects consumer biases can have on the substantive consequences of their tax provisions.

Conclusion

Customer-based incentives are an innovative, if somewhat peculiar, way to provide firms with targeted economic development incentives. However, this Article has demonstrated that, despite their peculiarity, customer-based incentives may be better suited than traditional incentives to achieve the substantive goals of states attempting to lure mobile firms. Not only do customer-based incentives have the potential to be more effective and to generate more social welfare than traditional incentives, they also appear better able to address equitable issues that traditional incentives face. As firms and states come to better understand customer-based incentives, their use should be expected to grow.

Taking a step back, these conclusions indicate that the form of tax provisions can have significant substantive effects, providing policymakers with opportunities to improve the effectiveness of tax provisions, particularly tax relief provisions. The lessons of behavioral research suggest that tax relief is most effective when it affects—even only nominally—a larger number of taxpayers, even if the substantive policy behind the relief is to influence only a small number of taxpayers. Additionally, nominally spreading tax relief among a broader population may appear more equitable to people, particularly when the people perceive they have some control over receiving the tax relief. Some may prefer to hide their taxes204 but keep their tax relief in plain sight.

• 1See John Buhl, New Jersey Governor, Amazon Reach Collection Agreement, 64 State Tax Notes 676, 676 (2012).
• 2See Peter D. Enrich, Saving the States from Themselves: Commerce Clause Constraints on State Tax Incentives for Business, 110 Harv L Rev 377, 382–84 (1996) (offering a brief history of state economic development incentives).
• 3See generally Billy Hamilton, Amazon’s “Dirty Little Secret”, 72 State Tax Notes 531 (2014); Amy Hamilton, Connecticut Governor Announces Deal with Amazon, State Tax Today 24­4 (Feb 5, 2013); Neil Downing, Massachusetts Governor, Amazon Announce Sales Tax Collection Deal, State Tax Today 239­7 (Dec 12, 2012); Billy Hamilton, What an Amazon Sales Tax Deal Looks Like, 65 State Tax Notes 675 (2012); John Buhl, Texas Announces Tax Collection Deal with Amazon, 64 State Tax Notes 351 (2012); John Buhl, Amazon Reaches Tax Deal with Nevada, Negotiating with Texas, 64 State Tax Notes 273 (2012); John Buhl, Virginia Governor Approves Amazon Sales Tax Deal, 64 State Tax Notes 85 (2012); Tom Humphrey, Tennessee Governor Approves Amazon Collection Exemption, 64 State Tax Notes 9 (2012); John Buhl, Indiana Governor: Amazon Will Collect Sales Tax, 63 State Tax Notes 191 (2012); Simon Brown, South Carolina Sales Tax Collection Exemption for Amazon Becomes Law, State Tax Today 111-40 (June 9, 2011).
• 4See Enrich, 110 Harv L Rev at 382–84 (cited in note 2).
• 5Luke Broadwater, Amazon’s Baltimore Site to Receive More than $43M in Tax Credits (Baltimore Sun, Oct 23, 2013), archived at http://perma.cc/2PRE-BHUW.
• 6Professor Edward J. McCaffery is a pioneer of this line of thinking about taxes and has argued that taxpayer behavior should be expected to affect tax systems. See Edward J. McCaffery, Cognitive Theory and Tax, 41 UCLA L Rev 1861, 1867–68 (1994). This Article builds on his and others’ early works by focusing on how consumer biases affect the substantive impact of tax relief provisions. See generally id.
• 7See James K. Smith, Use of Business Tax Incentives: Part 1, 17 J State Taxn 1, 9–15 (1999); Kathryn A. Pischak, State Economic Development Incentives: What’s Available? What Works?, 8 J State Taxn 191, 192 (1989).
• 8See Part I.A.
• 9See Part II.
• 10See Part II.A.
• 11See Deborah H. Schenk, Exploiting the Salience Bias in Designing Taxes, 28 Yale J Reg 253, 261–62 (2011); David Gamage and Darien Shanske, Three Essays on Tax Salience: Market Salience and Political Salience, 65 Tax L Rev 19, 23 (2011); Brian Galle, Hidden Taxes, 87 Wash U L Rev 59, 62 (2009).
• 12See, for example, Raj Chetty, Adam Looney, and Kory Kroft, Salience and Taxation: Theory and Evidence, 99 Am Econ Rev 1145, 1165 (2009).
• 13Lilian V. Faulhaber, The Hidden Limits of the Charitable Deduction: An Introduction to Hypersalience, 92 BU L Rev 1307, 1317 (2012).
• 14See Gamage and Shanske, 65 Tax L Rev at 49–50 (cited in note 11).
• 15See Abigail B. Sussman and Christopher Y. Olivola, Axe the Tax: Taxes Are Disliked More Than Equivalent Costs, 48 J Mktg Rsrch S91, S100 (2011).
• 16Reports indicate that Amazon has been the initial party proposing the customer-based incentives. See John Buhl, Amazon Seeking New Jersey Tax Collection Exemption, 63 State Tax Notes 511, 511 (2012); John Buhl, Amazon Offers Florida Jobs in Exchange for Collection Break, 63 State Tax Notes 360, 360 (2012); John Buhl, Amazon Reportedly Seeking Sales Tax Collection Exemption from South Carolina, State Tax Today 41-20 (Mar 2, 2011); Amy Hamilton, Amazon Proposes Deal in California, State Tax Today 171-4 (Sept 2, 2011).
• 17See Part III.B.
• 18See Part IV.A.
• 19See Part IV.A.
• 20See Part IV.B.
• 21See Part IV.B.
• 22See Part IV.C.
• 23See Part IV.C.
• 24See Randle B. Pollard, “Was the Deal Worth It?”: The Dilemma of States with Ineffective Economic Incentives Programs, 11 Hastings Bus L J 1, 3–8 (2015) (offering a history of state economic development incentives); Enrich, 110 Harv L Rev at 382–89 (cited in note 2). States also have a history of providing general economic development incentives, which are incentives available to all who meet certain criteria, such as performing particular levels of manufacturing in the state. See Pischak, 8 J State Taxn at 192–93 (cited in note 7).
• 25See Broadwater, Amazon’s Baltimore Site (cited in note 5).
• 26See Pollard, 11 Hastings Bus L J at 8–12 (cited in note 24); Kirk J. Stark and Daniel J. Wilson, What Do We Know about the Interstate Economic Effects of State Tax Incentives?, 4 Georgetown J L & Pub Pol 133, 137–41 (2006); Daniel P. Petrov, Note, Prisoners No More: State Investment Relocation Incentives and the Prisoners’ Dilemma, 33 Case W Reserve J Intl L 71, 72 (2001); Steven R. Little, Comment, Corporate Welfare Wars: The Insufficiency of Current Constraints on State Action and the Desirability of a Federal Legislative Response, 22 Hamline L Rev 849, 855–56 (1999); Smith, 17 J State Taxn at 9–15 (cited in note 7); Pischak, 8 J State Taxn at 192 (cited in note 7).
• 27See Pollard, 11 Hastings Bus L J at 13 (cited in note 24); Dale A. Oesterle, State and Local Government Subsidies for Businesses: A Siren’s Trap, 6 Ohio St Entrepreneurial Bus L J 491, 494 (2011); Zachary A. Phelps, Note, Stadium Construction for Professional Sports: Reversing the Inequities through Tax Incentives, 18 St John’s J Legal Commen 981, 1011 (2004); Sherry L. Jarrell, Gary Shoesmith, and J. Neal Robbins, Law and Economics of Regulating Local Economic Development Incentives, 41 Wake Forest L Rev 805, 822 (2006); Petrov, Note, 33 Case W Reserve J Intl L at 72–73 (cited in note 26); Matthew Schaefer, State Investment Attraction Subsidy Wars Resulting from a Prisoner’s Dilemma: The Inadequacy of State Constitutional Solutions and the Appropriateness of a Federal Legislative Response, 28 NM L Rev 303, 308 (1998); Smith, 17 J State Taxn at 17–18 (cited in note 7); Kimberly Galligan Key and James K. Smith, Trends in State and Local Economic Development Incentives, 15 J State Taxn 1, 2–3 (1996); Paul Stephen Dempsey, Primary Tax Incentives for Industrial Investment in the Southeastern United States, 25 Emory L J 789, 794–95 (1976).
• 28See, for example, Oesterle, 6 Ohio St Entrepreneurial Bus L J at 495–96 (cited in note 27) (noting skepticism about the effectiveness of economic development incentives); Petrov, Note, 33 Case W Reserve J Intl L at 77–79 (cited in note 26) (detailing the debate over the effectiveness of economic development incentives); Schaefer, 28 NM L Rev at 309–11 (cited in note 27) (same); Smith, 17 J State Taxn at 15–21 (cited in note 7) (same).
• 29See James R. Rogers, The Law and Policy of State Tax Competition: Much Ado about Nothing?, 4 Georgetown J L & Pub Pol 101, 105–07 (2006); Petrov, Note, 33 Case W Reserve J Intl L at 79–83 (cited in note 26); Schaefer, 28 NM L Rev at 311–12 (cited in note 27); Enrich, 110 Harv L Rev at 391–97 (cited in note 2); James R. Rogers, State Tax Competition and Congressional Commerce Power: The Original Prudence of Concurrent Taxing Authority, 7 Regent U L Rev 103, 108–13 (1996).
• 30See text accompanying note 26.
• 31States impose sales taxes on sales of taxable property and services. See, for example, Cal Rev & Tax Code § 6051; NY Tax Law § 1105; Tex Tax Code Ann § 151.051; Wis Stat § 77.52. Some states impose their sales taxes directly on consumers; others impose the taxes directly on vendors. Compare NY Tax Law § 1133 (imposing liability for the tax on the consumer) and 20 NYCRR § 525.2(a)(4) (explaining the nature of the sales tax as a “consumer tax”), with Cal Rev & Tax Code § 6051 (imposing liability for the tax on the vendor). In either case, vendors are required or permitted to collect sales taxes from their customers at the point of sale; if a vendor fails to do so, it becomes legally responsible for the uncollected taxes if it was not already so responsible. See, for example, Cal Civ Code § 1656.1; Fla Stat § 212.07; NY Tax Law § 1131(1); NC Gen Stat § 105-164.8; Tex Tax Code Ann § 151.052; Utah Code Ann § 59-12-107(2)(a).

  Additionally, states impose use taxes on goods legally subject to sales tax but on which sales tax was not paid. See Richard D. Pomp, State & Local Taxation at 6-40 to -42 (Richard D. Pomp 8th ed 2015). This situation might occur when a consumer purchases a good that would have been taxable in her home state in a sales-tax-free state, such as Delaware, and then brings the good back home for use. See id at 6-39. Use taxes are imposed on the consumer, but states allow credits against their use taxes for sales taxes properly paid to other jurisdictions on the sale of taxable goods in order to prevent double taxation of the goods. See id. As with sales taxes, vendors are required to collect use taxes at the point of sale. See, for example, Cal Rev & Tax Code § 6203; Fla Stat § 212.07; NY Tax Law § 1131(1); NC Gen Stat § 105-164.8; Tex Tax Code Ann § 151.103; Utah Code Ann § 59-12-107(2)(a). Given that sales and use taxes are imposed on a transaction-by-transaction basis, the vendor-collection scheme is viewed as the most administratively effective approach to collecting the taxes. See John A. Swain, State Sales and Use Tax Jurisdiction: An Economic Nexus Standard for the Twenty-First Century, 38 Ga L Rev 343, 345 (2003) (“As between collecting tax from each individual consumer or from the seller, it is more administratively practical to collect the tax from the seller.”).

• 32From a representative-government perspective, the state is largely an agent of the customers, but because state actors can have independent motivations and some customers may not also be voters, the state and the customers are approached here as separate parties.
• 33All else being equal between traditional incentives and customer-based incentives, these administrative savings should cause a firm that does not currently have a physical presence in the taxing state to prefer customer-based incentives, as the firm will thus be able to continue with its status quo position of not collecting sales taxes in the state even though it would have a physical presence in the state. Many thanks to Dhammika Dharmapala for raising this point. A corollary to this point is that a firm experiencing the benefits of not collecting taxes because it lacks a physical presence in the offering state, see note 34, would prefer to retain those benefits upon entering the state, making customer-based incentives more appealing than traditional incentives.
• 34Under Supreme Court precedent, a vendor must have a physical presence in a state before that state may require it to collect or pay sales tax. See Quill Corp v North Dakota, 504 US 298, 317 (1992). The boundaries of what constitutes a physical presence in a state are not completely clear, though the Supreme Court has declared that the furthest extension of the concept has been to attribute to an out-of-state vendor the in-state presence of a person acting on the vendor’s behalf to establish or maintain a market for the vendor in the state. See id at 306. See also Tyler Pipe Industries, Inc v Washington State Department of Revenue, 483 US 232, 249–50 (1987); Scripto, Inc v Carson, 362 US 207, 209–11 (1960). In the absence of additional guidance, the states have introduced various standards outlining their views of the limits of the physical-presence standard. See Pomp, State & Local Taxation at 9-71 to -74 (cited in note 31).
• 35See Jeffrey L. Hoopes, Jacob R. Thornock, and Braden M. Williams, Does Use Tax Evasion Provide a Competitive Advantage to E-tailers?, 69 Natl Tax J 133, 163–64 (2016) (summarizing evidence regarding the competitive advantage gained by online vendors who do not have to collect sales taxes); Liran Einav, et al, Sales Taxes and Internet Commerce, 104 Am Econ Rev 1, 3 (2014) (“We estimate that on average, the application of a 10 percent sales tax [on the vendor] reduces purchases by 15 percent among [eBay] buyers who have clicked on an item.”); Yu Jeffrey Hu and Zhulei Tang, The Impact of Sales Tax on Internet and Catalog Sales: Evidence from a Natural Experiment, 32 Intl J Indust Org 84, 90 (2014) (providing empirical support for the conclusion that “a change in sales tax has a highly significant and large effect on remote sales through the internet and catalog channels”); Brian Baugh, Itzhak Ben-David, and Hoonsuk Park, Can Taxes Shape an Industry? Evidence from the Implementation of the “Amazon Tax” *24 (Fisher College of Business Working Paper No 2014-03-05, Sept 1, 2016), archived at http://perma.cc/RUG5-U9NG (demonstrating that consumers prefer to make purchases from firms that do not collect sales taxes); Eric T. Anderson, et al, How Sales Taxes Affect Customer and Firm Behavior: The Role of Search on the Internet, 47 J Mktg Rsrch 229, 239 (2010) (finding that an obligation to collect sales taxes decreases Internet orders from a retailer by 11.6 percent, though finding no similar effect on catalog orders); Joel Slemrod, Does It Matter Who Writes the Check to the Government? The Economics of Tax Remittance, 61 Natl Tax J 251, 254 (2008) (describing the benefit of a firm not remitting taxes when its competitors do remit).
• 36Such language directly providing the benefit to the firm might be unconstitutional in many states in which tax laws are required to apply to all people equally, so some creative drafting might be required. For an example of such creativity, see SC Code Ann § 12-36-2691 (providing indirect customer-based incentives to Amazon without nominally singling out Amazon). See also Clif LeBlanc, House OKs Amazon Deal (The State, May 18, 2011), archived at http://perma.cc/WU3N-7X5V (documenting the conclusion of tax incentive negotiations between Amazon and South Carolina lawmakers, which were focused on job creation and investment targets).
• 37See, for example, 35 ILCS 120/2-5(35-5) (imposing a lower tax rate than the general tax rate on “food for human consumption that is to be consumed off the premises where it is sold”).
• 38See, for example, Ind Code § 6-2.5-5-26 (providing exemptions for certain sales by nonprofit organizations).
• 39See, for example, NJ Rev Stat § 52:27H-80 (providing an exemption from half of the tax for sales by qualifying businesses located in a designated enterprise zone).
• 40See Kyle R. Williams, Note, State Tax Credits for Private Start-Up Capital: Arching toward Urban “Entrepreneurial Redevelopment”, 6 Wash U J L & Pol 299, 315 (2001); Smith, 17 J State Taxn at 12 (cited in note 7); Pischak, 8 J State Taxn at 194 (cited in note 7).
• 41In addition, the insights gained from the study of customer-based incentives may help policymakers craft more effective enterprise zones, specifically if future enterprise zones are able to harness the potential benefits created by consumer behaviors discussed in this Article. Research on the effectiveness of enterprise zones finds that the zones often struggle to meet their goals. See, for example, David Neumark and Jed Kolko, Do Enterprise Zones Create Jobs? Evidence from California’s Enterprise Zone Program, 68 J Urban Econ 1, 15 (2010) (concluding that California’s enterprise zones failed to increase employment); Joel A. Elvery, The Impact of Enterprise Zones on Resident Employment: An Evaluation of the Enterprise Zone Programs of California and Florida, 23 Econ Development Q 44, 57 (2009) (concluding that enterprise zones in California and Florida had no significant effect on employment); Stephen Billings, Do Enterprise Zones Work? An Analysis at the Borders, 37 Pub Fin Rev 68, 89 (2009) (finding that Colorado’s enterprise zones had insignificant impacts on firms’ location decisions but had positive impacts for job creation).
• 42See Nina Manzi, Use Tax Collection on Income Tax Returns in Other States *10 (Minn House Rsrch Department, April 2015), archived at http://perma.cc/J4W3-UGM6 (noting that the percentage of taxpayers who report use tax in states in which that tax can be reported on income-tax returns is approximately 1.9 percent). The states’ inability to collect sales taxes from consumers is grounded in administrative and political difficulties. See Adam B. Thimmesch, Taxing Honesty, 118 W Va L Rev 147, 151–60 (2015); Swain, 38 Ga L Rev at 353 (cited in note 31); Walter Hellerstein, Jurisdiction to Tax Income and Consumption in the New Economy: A Theoretical and Comparative Perspective, 38 Ga L Rev 1, 23–24 (2003). Business consumers, as opposed to individual consumers, have higher compliance rates with their sales tax obligations. See Thimmesch, 118 W Va L Rev at 158 (cited in note 42).
• 43See note 34. In addition to the physical-presence standard, which is imposed under the Supreme Court’s interpretation of the Commerce Clause, the Due Process Clause also places limits on the states’ jurisdiction to tax remote vendors, though the scope of those limitations is debated. Compare Hayes R. Holderness, Taking Tax Due Process Seriously: The Give and Take of State Taxation, 20 Fla Tax Rev 371, 419–25 (2017) (arguing that the Due Process Clause places meaningful limitations on states’ abilities to tax remote vendors making sales into the states), with Michael T. Fatale, The Evolution of Due Process and State Tax Jurisdiction, 55 Santa Clara L Rev 565, 586–92 (2015) (arguing that the Due Process Clause does not place significant limitations on states’ abilities to tax remote vendors making sales into the states).
• 44Maintaining the status quo of noncollection may be a primary driver of a remote vendor, such as Amazon, seeking customer-based incentives rather than traditional incentives. See note 33.
• 45The states’ discontent with this result has led them to campaign against the physical-presence standard, with little success so far. See generally, for example, Maria Koklanaris, Governors: States Will Act on Their Own for E-Fairness, 79 State Tax Notes 119 (2016) (reporting on states’ efforts to overturn Quill); Brian Bardwell, Council of State Governments Asks Congress to Act on E-Commerce Taxation, 79 State Tax Notes 27 (2016) (same); Jennifer DePaul, Governors Press for Passage of MFA, 75 State Tax Notes 79 (2015) (same); David Brunori, MTC Market-Based Sourcing Efforts Are Good, 78 State Tax Notes 915 (2015) (“[I]f Congress does not act, there is a good possibility that the Supreme Court will overturn Quill. States are getting very aggressive regarding sales tax nexus. More litigation is coming.”). See also Annette Nellen, Still Seeking Digital Direction, 78 State Tax Notes 797, 797–98 (2015) (discussing Alabama’s efforts to overturn Quill); Adam B. Thimmesch, The Fading Bright Line of Physical Presence: Did KFC Corporation v. Iowa Department of Revenue Give States the Secret Recipe for Repudiating Quill?, 100 Ky L J 339, 340 (2011–12) (“States have responded to these losses by aggressively and continuously lobbying Congress to legislatively overturn the physical-presence rule. Despite those efforts, however, Congress has not yet given states the reprieve that they seek.”); Swain, 38 Ga L Rev at 370 (cited in note 31) (“Unfortunately, Congress has not proven to be an effective forum for state tax reform. Ever since Bellas Hess was decided in 1967, legislation that would ‘overrule’ the physical presence test has been introduced, only to flounder.”).

  However, even though the standard remains in place, momentum appears to be gaining against it: Congress has considered legislation to overturn the standard, advocates have called for the Supreme Court to overturn the standard, and some states have gone so far as to adopt legislation or regulations that intentionally ignore the physical-presence standard. See Marketplace Fairness Act of 2017, S 976, 115th Cong, 1st Sess (Apr 27, 2017); SD Cod Laws § 10-64-1 et seq; Ala Reg § 810-6-2-.90.03; Tenn Comp R & Regs 1320-05-01-.129; David Brunori, It’s Time to Overturn Quill, 55 State Tax Notes 497, 497–98 (2010) (advocating for the overturning of Quill); Robert D. Plattner, Quill: Ten Years After, 25 State Tax Notes 1017, 1017 (2002):

  [T]he Quill decision qualifies as a blunder of major proportions by the Court. . . . [T]he states should push the Supreme Court to reexamine Quill by bringing a new test case that seeks to change not only the outcome in Quill but also the framework of Supreme Court decisionmaking in state tax nexus cases. While it may be naive to think that the Court would abandon Quill, it is hard to believe that the Supreme Court is satisfied with the anachronistic, illogical state of constitutional doctrine embodied in Quill. Perhaps, given another opportunity to do better, the Court would seize on it.

  Litigation has already begun in South Dakota and Alabama over their efforts to require remote vendors to collect sales taxes. See generally South Dakota v Wayfair, Inc, 229 F Supp 3d 1026 (D SD 2017); Statement of Newegg Inc. in Support of Its Notice of Appeal to the Alabama Tax Tribunal, Newegg Inc v State of Alabama Department of Revenue, Docket No S 16-613 (Ala Tax Trib filed Aug 26, 2016). See also generally Maria Koklanaris, Both Sides Pleased With Court Ruling Striking Down South Dakota Remote Sales Tax Law, 2017 State Tax Today 44-2 (Mar 8, 2017) (noting South Dakota representatives’ desire to take their case to the US Supreme Court); Maria Koklanaris, Retailer Challenges Alabama’s Economic Nexus Rule, 80 State Tax Notes 918 (2016).

• 46See generally Donald Bruce, William F. Fox, and LeAnn Luna, State and Local Sales Tax Revenue Losses from E-Commerce, 52 State Tax Notes 537 (2009). The researchers expect that such revenue losses “will likely continue to grow rapidly, at least for the next several years.” Donald Bruce, William F. Fox, and LeAnn Luna, E-tailer Sales Tax Nexus and State Tax Policies, 68 Natl Tax J 735, 736 (2015). Some commentators have questioned the accuracy of this estimate. See, for example, Joseph Henchman, The Marketplace Fairness Act: A Primer (Tax Foundation, July 14, 2014), archived at http://perma.cc/CU6X-Z6W9; Billy Hamilton, Fox and Friends: The Rest of the Story on E-Commerce Tax Loss Estimates, 68 State Tax Notes 535, 535–39 (2013); Noah Aldonas, DOR Disputes E-Commerce Sales Tax Loss Estimates, 65 State Tax Notes 576, 576 (2012).
• 47See note 45.
• 48See note 3. Though they could theoretically do so, these states have not provided direct sales tax relief to Amazon’s customers, but as explained in Part I.A, those customers receive sales tax relief indirectly due to the states’ agreements with Amazon.
• 49It is no surprise that Amazon is the firm that introduced customer-based incentives to the world. Consider the following: Traditionally, targeted economic development incentives were provided to manufacturing firms, such as airplane and car makers, which employed large numbers of people and required large investments in infrastructure. See generally Billy Hamilton, 16 Things You Need to Know about State Tax Breaks, 74 State Tax Notes 267 (2014). See also Peter K. Eisinger, The Rise of the Entrepreneurial State 78–81 (Wisconsin 1988). Targeted customer-based incentives would have been unintuitive to craft for manufacturers, as manufacturers typically do not have to collect taxes from their customers. Thus, traditional incentives arose in the forms that they did, such as income tax credits and property tax abatements. As the US economy shifted from manufacturing to services, relatively fewer large manufacturers were available for states to lure. See Joshua P. Rubin, Note, Take the Money and Stay: Industrial Location Incentives and Relational Contracting, 70 NYU L Rev 1277, 1306 (1995) (noting the decline in manufacturing jobs). The states had to pay more for those firms or shift their efforts elsewhere. States, accustomed to offering targeted traditional incentives to manufacturers, began offering the same types of incentives to vendors without considering the possibility of customer-based incentives. However, when Amazon began soliciting targeted economic development incentives, its history of not having to collect sales taxes came into play—for years it promoted the sales-tax-(collection)-free shopping experience its customers could have by purchasing from it. See Pomp, State & Local Taxation at 6-41 (cited in note 31). Realizing the value of the competitive advantage of not collecting sales taxes, Amazon asked for customer-based incentives, introducing the concept to state policymakers. See note 16. This series of events, or something similar, would explain how Amazon came to be the first mover in this area.
• 50See David Brunori, Requiem for Good Sales Tax Policy, 61 State Tax Notes 63, 63 (2011) (“By letting one of the world’s largest retailers off the hook for collecting the tax, Texas would have changed how everyone views the sales tax system. Other large retailers—those with the requisite lobbying resources—would have sought similar deals.”).
• 51Einav, et al, 104 Am Econ Rev at 24 (cited in note 35). See also note 35.
• 52Indeed, Amazon’s customer-based incentives have typically been for the relief of its tax collection obligations for a one- or two-year period. See note 3.
• 53See Baugh, Ben-David, and Park, Can Taxes Shape an Industry? at *24 (cited in note 35); Hamilton, 72 State Tax Notes at 531 (cited in note 3); Henry J. Reske, Amazon Study: Customers May Look Elsewhere When Sales Tax Advantage Ends, Sales Tax Today 78-1 (2014).
• 54Even if the state places a limit on the sales taxes it will forego through the customer-based incentives, customer-based incentives may produce larger benefits for the firm than traditional incentives because of the way consumers react to each type of incentive. See Part III.
• 55See, for example, IRC § 30B. However, one study found that sales tax incentives were more effective than income tax incentives at generating demand for hybrid vehicles. See Kelly Sims Gallagher and Erich Muehlegger, Giving Green to Get Green? Incentives and Consumer Adoption of Hybrid Vehicle Technology, 61 J Envir Econ & Mgmt 1, 9 (2011).
• 56Alternatively, if consumers control prices, then they could require that prices fall by the full amount of the incentives, capturing the incentives.
• 57Jonathan Gruber, Public Finance and Public Policy 560–62 (Worth 2d ed 2007).
• 58See Raymond E. Owens and Pierre-Daniel Sarte, Analyzing Firm Location Decisions: Is Public Intervention Justified?, 86 J Pub Econ 223, 224 (2002) (making a similar assumption in the context of targeted economic development incentives: “[s]ince the typical firm concerned is relatively large, and often faces a national demand curve for its product, we allow firms to act as price setters”).
• 59See, for example, R.H. Coase, The Problem of Social Cost, 3 J L & Econ 1, 2–6 (1960); Rubert Sausgruber and Jean-Robert Tyran, Tax Salience, Voting, and Deliberation *4, 6 (University of Copenhagen Department of Econ Discussion Paper No 08-21, Oct 2008), archived at http://perma.cc/487P-P962 (describing “Tax Liability Side Equivalence,” the principle that “it is matter [sic] of indifference whether a tax is levied on the buyers or sellers in a market,” as a “fundamental principle in public economics”); Slemrod, 61 Natl Tax J at 255–56 (cited in note 35) (describing the “Theorem of the Invariance of Tax Incidence”).
• 60See Richard H. Thaler, Mental Accounting and Consumer Choice, 27 Mktg Sci 15, 25 (2008) (discussing fairness concerns); Cass R. Sunstein, Behavioral Analysis of Law, 64 U Chi L Rev 1175, 1179–81 (1997) (discussing loss aversion).
• 61See, for example, Russell B. Korobkin and Thomas S. Ulen, Law and Behavioral Science: Removing the Rationality Assumption from Law and Economics, 88 Cal L Rev 1051, 1136 (2000); Christine Jolls, Cass R. Sunstein, and Richard Thaler, A Behavioral Approach to Law and Economics, 50 Stan L Rev 1471, 1479, 1489–96 (1998); Sunstein, 64 U Chi L Rev at 1186 (cited in note 60); Daniel Kahneman, Jack L. Knetsch, and Richard H. Thaler, Fairness and the Assumptions of Economics, 59 J Bus S285, S288–92 (1986); Daniel Kahneman, Jack L. Knetsch, and Richard H. Thaler, Fairness as a Constraint on Profit Seeking: Entitlements in the Market, 76 Am Econ Rev 728, 736–37 (1986).
• 62See Korobkin and Ulen, 88 Cal L Rev at 1135–36 (cited in note 61) (discussing the ultimatum game); Jolls, Sunstein, and Thaler, 50 Stan L Rev at 1489–93 (cited in note 61) (describing findings of ultimatum game experiments).
• 63Jolls, Sunstein, and Thaler, 50 Stan L Rev at 1489–90 (cited in note 61).
• 64Id.
• 65Id at 1490.
• 66See id at 1490–93.
• 67See Kahneman, Knetsch, and Thaler, 76 Am Econ Rev at 736 (cited in note 61) (“[A] history or reputation of unfair dealing may induce potential transactors to take their business elsewhere.”).
• 68See Jolls, Sunstein, and Thaler, 50 Stan L Rev at 1515 (cited in note 61) (“[R]ecent evidence of price stickiness shows that firms’ behavior seems to be affected greatly by their customers’ perceptions of unfair price increases.”); Kahneman, Knetsch, and Thaler, 59 J Bus at S287 (cited in note 61).
• 69Kahneman, Knetsch, and Thaler, 76 Am Econ Rev at 729 (cited in note 61).
• 70Kahneman, Knetsch, and Thaler, 59 J Bus at S296 (cited in note 61); Kahneman, Knetsch, and Thaler, 76 Am Econ Rev at 729–30 (cited in note 61). See also Jolls, Sunstein, and Thaler, 50 Stan L Rev at 1496 (cited in note 61).
• 71Kahneman, Knetsch, and Thaler, 59 J Bus at S297 (cited in note 61).
• 72Id at S296 (“In a conflict between the transactor’s claim to the reference price . . . and the firm’s claim to its reference profit, it is acceptable for the firm to impose its claim rather than compromise.”); Kahneman, Knetsch, and Thaler, 76 Am Econ Rev at 732 (cited in note 61).
• 73Kahneman, Knetsch, and Thaler, 59 J Bus at S293–96 (cited in note 61); Kahneman, Knetsch, and Thaler, 76 Am Econ Rev at 734–36 (cited in note 61).
• 74Kahneman, Knetsch, and Thaler, 59 J Bus at S293–94 (cited in note 61); Kahneman, Knetsch, and Thaler, 76 Am Econ Rev at 734 (cited in note 61).
• 75See Gamage and Shanske, 65 Tax L Rev at 29 (cited in note 11) (“[T]axpayers appear to spotlight on the prices charged (or displayed) at the time of market decision-making.”); Schenk, 28 Yale J Reg at 266 (cited in note 11) (“[S]tudies confirm the intuition that unless the amount of the tax is prominent, consumers will make consumption decisions ‘pretax,’ even though they are aware that a tax will be imposed and even where calculation of the tax is relatively uncomplicated.”).
• 76See generally, for example, Andrew T. Hayashi, The Legal Salience of Taxation, 81 U Chi L Rev 1443 (2014); Brian Galle, Carrots, Sticks, and Salience, 67 Tax L Rev 53 (2013); Faulhaber, 92 BU L Rev 1307 (cited in note 13); Gamage and Shanske, 65 Tax L Rev 19 (cited in note 11); Schenk, 28 Yale J Reg 253 (cited in note 11); Jacob Nussim, To Confuse and Protect: Taxes and Consumer Protection, 1 Colum J Tax L 218 (2010); Galle, 87 Wash U L Rev at 59 (cited in note 11); Sausgruber and Tyran, Tax Salience (cited in note 59).
• 77See Part III.A. See also, for example, Gamage and Shanske, 65 Tax L Rev at 26–54 (cited in note 11) (surveying tax salience research); Chetty, Looney, and Kroft, 99 Am Econ Rev at 1145 (cited in note 12).
• 78Schenk, 28 Yale J Reg at 262 (cited in note 11); Gamage and Shanske, 65 Tax L Rev at 24 (cited in note 11); Galle, 87 Wash U L Rev at 62 (cited in note 11).
• 79The literature on tax salience typically splits the term into market salience and “political salience”—not practical salience. See Gamage and Shanske, 65 Tax L Rev at 24 (cited in note 11); Schenk, 28 Yale J Reg at 272 (cited in note 11). I use the term “practical salience” because “political salience” as used in the literature often focuses on the type of awareness that leads to political reactions, which may not fully cover the awareness that a tax provision is the source of economic effects, which is the concept I wish to describe. Market salience and practical—or political—salience are not the only types of tax salience that a taxpayer might experience. As an example of another type of salience that might accompany a tax provision, Professor Andrew Hayashi argues that a provision’s “legal salience”—“the visibility or prominence of a tax [that] affects the likelihood that taxpayers will use administrative remedies to reduce [the tax’s] burden”—impacts the distribution of the burden of the tax provision. Hayashi, 81 U Chi L Rev at 1456–58 (cited at note 76).
• 80See David Gamage, On the Future of Tax Salience Scholarship: Operative Mechanisms and Limiting Factors, 41 Fla St U L Rev 173, 180 (2013) (“[B]oth market salience and political salience are likely to result from similar operative mechanisms.”); Schenk, 28 Yale J Reg at 273–76 (cited in note 11) (discussing the factors that influence the two types of salience); Gamage and Shanske, 65 Tax L Rev at 54–58 (cited in note 11).
• 81Chetty, Looney, and Kroft, 99 Am Econ Rev at 1158–64 (cited in note 12).
• 82Id.
• 83See note 79.
• 84See Gamage and Shanske, 65 Tax L Rev at 24–25 (cited in note 11); Schenk, 28 Yale J Reg at 273–74 (cited in note 11).
• 85See Gamage and Shanske, 65 Tax L Rev at 34 (cited in note 11) (“[I]t has been notoriously difficult to test hypotheses for political salience.”).
• 86See Schenk, 28 Yale J Reg at 276, 285 (cited in note 11). See also Edward J. McCaffery and Jonathan Baron, Thinking about Tax, 12 Psychology, Pub Pol & L 106, 107–08 (2006) (“People decide complex matters—and tax raises a host of complex matters—by responding to the most salient or obvious aspect of a choice set or decision problem. They fail to take into account logically relevant information that is not immediately available to their mental models.”).
• 87See Gamage and Shanske, 65 Tax L Rev at 39–41 (cited in note 11); Schenk, 28 Yale J Reg at 272, 277 (cited in note 11); Aradhna Krishna and Joel Slemrod, Behavioral Public Finance: Tax Design as Price Presentation, 10 Intl Tax & Pub Fin 189, 193–94 (2003) (discussing the effects of tax withholding).
• 88See Gamage, 41 Fla St U L Rev at 184 (cited in note 80) (“The pennies-a-day concept thus supports the tax-system-complexity hypothesis that tax instruments which levy many smaller payments over a period of time (e.g., sales taxes) may have lower political salience than tax instruments which levy fewer larger tax payments (e.g., property taxes).”); Krishna and Slemrod, 10 Intl Tax & Pub Fin at 198 (cited in note 87) (discussing the effects of tax obfuscation); Schenk, 28 Yale J Reg at 256–61, 277–80 (cited in note 11) (discussing transparency and complexity in the political process regarding taxes); Joel Slemrod, Old George Orwell Got It Backward: Some Thoughts on Behavioral Tax Economics, 66 FinanzArchiv/Pub Fin Analysis 15, 18–21 (2010) (discussing voter confusion and its effect on tax complexity).
• 89See Gamage and Shanske, 65 Tax L Rev at 35–38 (cited in note 11); Schenk, 28 Yale J Reg at 727 (cited in note 11); Edward J. McCaffery and Jonathan Baron, The Political Psychology of Redistribution, 52 UCLA L Rev 1745, 1761–65 (2005) (discussing the effects of indirect “hidden” taxes); Sausgruber and Tyran, Tax Salience at *2 (cited in note 59); Krishna and Slemrod, 10 Intl Tax & Pub Fin at 190 (cited in note 87).
• 90See Schenk, 28 Yale J Reg at 274–76 (cited in note 11) (describing how various forms of taxation might have various levels of both types of salience); Gamage and Shanske, 65 Tax L Rev at 54–58 (cited in note 11).
• 91See Chetty, Looney, and Kroft, 99 Am Econ Rev at 1158–64 (cited in note 12). See also Gamage and Shanske, 65 Tax L Rev at 56 (cited in note 11).
• 92See generally, for example, Philip Mattera, Kasia Tarczynska, and Greg LeRoy, Tax Breaks and Inequality: Enriching Billionaires and Low-Road Employers in the Name of Economic Development (Good Jobs First, Dec 2014), archived at http://perma.cc/3SDC-3QUY. Good Jobs First is a prime example of a watchdog group that eyes targeted economic development incentives. The group’s “Subsidy Tracker” allows people access to information regarding the size of such incentives and what firms received them. See Subsidy Tracker 3.0 (Good Jobs First), archived at http://perma.cc/M3M2-FQKP.
• 93See note 88.
• 94As a firm is a not an individual, it does not itself have awareness of the tax relief. Thus, when discussing the firm’s economic reactions to the various incentives, it is more accurate to consider the reactions of the managers and owners of the firm (and customers, which are discussed in the text). See Brian Galle, Is Local Consumer Protection Law a Better Redistributive Mechanism than the Tax System?, 65 NYU Ann Surv Am L 525, 540 (2010) (“When a firm is liable for a judgment, the economic burden of that judgment is ultimately passed on to real people, whether they are the firm’s owners, its workers, its customers, or even investors in other businesses.”); Rogers, 4 Georgetown J L & Pub Pol at 109 (cited in note 29) (“[T]ax abatements shift money from governments to businesses, and that money is then passed along to consumers, workers and capital owners in the form of wages, lower prices, or higher capital returns.”). See also Gruber, Public Finance at 700 (cited in note 57); McCaffery, 41 UCLA L Rev at 1883 (cited in note 6).

  Because the tax relief from both forms of targeted economic development incentives is negotiated for by and is immediate to firm managers, the tax relief from both forms of incentives should have similar levels of practical salience to the managers. The existence of this practical salience is evidenced by the fact that managers seek out targeted economic development incentives from states. See generally, for example, Billy Hamilton, Down but Not Out: States Haven’t Given Up on Giveaways, 80 State Tax Notes 703 (2016) (discussing recent targeted economic development incentive offerings). Because customer-based incentives are new and not fully understood and their tax relief does not fall directly to firms, the tax relief from customer-based incentives may be less practically salient to managers than traditional incentives. The limited experience with customer-based incentives appears to bear this out; after all, only Amazon has sought out these incentives as of yet. See, for example, note 16. Even so, states wishing to provide customer-based incentives should be able to increase their practical salience to firms’ managers by educating the managers of their benefits. Additionally, because neither form of incentive directly affects owners of firms in their role as owners, the tax relief from neither form of incentive should be more or less practically salient to the owners. However, to the extent that the two forms of incentives are reported differently to owners (that is, reduced tax obligations due to traditional incentives versus more dispersed gains in profits due to customer-based incentives), the practical salience of the tax relief from the two forms of incentives may vary. Assuming though that owners’ primary concern is the profitability of the firm, the practical salience of the tax relief from the two different forms of incentives may not lead to meaningfully different reactions from owners to either form, provided that the relief is market salient. For a discussion on market salience of the relief to firms, see Part III.C. Further research into this question is needed to make any firm conclusions, however.

• 95See Gamage, 41 Fla St U L Rev at 190–91 (cited at note 80) (discussing the effects of loss aversion in the context of tax salience); Korobkin and Ulen, 88 Cal L Rev at 1111 (cited in note 61) (discussing the status quo bias); Sunstein, 64 U Chi L Rev at 1179–81 (cited in note 60) (discussing loss aversion); McCaffery, 41 UCLA L Rev at 1874–75 (cited in note 6) (same).
• 96See Joshua D. Rosenberg, The Psychology of Taxes: Why They Drive Us Crazy, and How We Can Make Them Sane, 16 Va Tax Rev 155, 173 (1996) (“The ex post enforcement of tax laws implicates a cognitive phenomenon known as loss aversion. Basically, people have a much greater aversion to giving up what we already have than we do to not getting what we do not yet have.”); Daniel Kahneman, Jack L. Knetsch, and Richard H. Thaler, Anomalies: The Endowment Effect, Loss Aversion, and Status Quo Bias, 5 J Econ Persp 193, 194–203 (Winter 1991); Jack L. Knetsch, The Endowment Effect and Evidence of Nonreversible Indifference Curves, 79 Am Econ Rev 1277, 1282–83 (1989); Tatiana A. Homonoff, Can Small Incentives Have Large Effects? The Impact of Taxes versus Bonuses on Disposable Bag Use *36–37 (Princeton University Working Paper No 575, Mar 27, 2013), archived at http://perma.cc/EWK8-A6TR (finding that a $0.05 tax on plastic bag use decreased usage by significantly more than equal bonuses).
• 97See Kahneman, Knetsch, and Thaler, 5 J Econ Persp at 195–97 (cited in note 96) (describing the mug experiment).
• 98See McCaffery and Baron, 12 Psychology, Pub Pol & L at 114–15 (cited in note 86).
• 99See Schenk, 28 Yale J Reg at 271–72 (cited in note 11).
• 100See Korobkin and Ulen, 88 Cal L Rev 1112 (cited in note 61) (observing evidence that default terms are likely to be seen as the status quo; thus those terms “will be ‘sticky’”). See also Knetsch, 79 Am Econ Rev at 1280 (cited in note 96) (describing people’s “strong aversion to giving up an initial entitlement”).
• 101See notes 78–79.
• 102The following analysis relies on the current state of research into the effects of tax salience, but this area of study is relatively new and much remains to be learned. See Schenk, 28 Yale J Reg at 270 (cited in note 11); Gamage and Shanske, 65 Tax L Rev at 23 (cited in note 11). Additionally, tax salience is unlikely to be the only thing affecting people’s behavior in any given situation. Schenk, 28 Yale J Reg at 270 (cited in note 11). Therefore, as the understanding of the effects of tax salience becomes more refined and additional variables are considered, the conclusions presented here may need adjustment. Even so, the following analysis is valuable because it provides a basis for considering the effects of tax salience on the effectiveness of tax provisions and demonstrates that the choice of form of a tax provision does matter even if other factors must be considered to fully understand how people’s behavior can influence the effectiveness of a tax provision. See Gamage and Shanske, 65 Tax L Rev at 59 (cited in note 11):

  It may thus be tempting to conclude that policy debates should ignore intuitions about tax salience until (or unless) these intuitions receive more satisfactory empirical support. Yet it must be recognized that intuitions about tax salience already significantly influence debates over tax policy. So long as important political actors (and perhaps also the voters on whose support they depend) make tax policy decisions based on naïve intuitions about tax salience, scholars must continue to analyze these intuitions based on whatever evidence can be mustered—no matter how inconclusive the evidence might be.

• 103See note 77.
• 104See Galle, 67 Tax L Rev at 63–67 (cited in note 76).
• 105See id at 64; Jacob Goldin, Note, Sales Tax Not Included: Designing Commodity Taxes for Inattentive Consumers, 122 Yale L J 258, 269 (2012); Faulhaber, 92 BU L Rev at 1309 (cited in note 13); Schenk, 28 Yale J Reg at 263 (cited in note 11); Nussim, 1 Colum J Tax L at 220 (cited in note 76); Chetty, Looney, and Kroft, 99 Am Econ Rev at 1165, 1175 (cited in note 12); Galle, 87 Wash U L Rev at 77 (cited in note 11).
• 106Galle, 67 Tax L Rev at 91–94 (cited in note 76); Faulhaber, 92 BU L Rev at 1317 (cited in note 13). Professor Lilian V. Faulhaber observes that hypersalience occurs “when a tax provision is fully—or almost fully—salient, but the limits restricting that provision’s application are hidden, or less salient.” Faulhaber, 92 BU L Rev at 1309 (cited in note 13). As developed, this Article argues that hypersalience can occur simply through an overestimation of taxes.
• 107See Gamage and Shanske, 65 Tax L Rev at 26 (cited in note 11). See also Edward J. McCaffery and Jonathan Baron, Isolation Effects and the Neglect of Indirect Effects of Fiscal Policies, 19 J Behav Dec Making 289, 289–91 (2006) (discussing isolation effects, which are similar to spotlighting). Spotlighting is not the only theoretical basis for differing levels of tax salience. Another popular theory suggests that “ironing” affects the salience of taxes. Ironing refers to the “smoothing” of tax rates by mentally applying one average tax rate to all income even though rates vary, causing a misunderstanding of one’s actual tax burden. See Gamage and Shanske, 65 Tax L Rev at 31–33 (cited in note 11) (discussing research on ironing). As Professors David Gamage and Darien Shanske note, “[i]n essence, ironing is a form of spotlighting behavior wherein taxpayers spotlight on their average tax rates instead of using their effective marginal tax rates.” Id at 31. In addition, the ironing theory depends on the availability of a complex schedule of tax rates and is thus more suited for understanding the salience of income taxes than sales taxes. For these reasons, this Article proceeds with the spotlighting theory to explain tax salience.
• 108Gamage, 41 Fla St U L Rev at 177 (cited in note 80).
• 109See Marika Cabral and Caroline Hoxby, The Hated Property Tax: Salience, Tax Rates, and Tax Revolts *29–30 (NBER Working Paper No 18514, Nov 2012), archived at http://perma.cc/N3P8-6ZKW (finding that automatically deducted property taxes were less salient than those paid manually); Amy Finkelstein, E-ZTax: Tax Salience and Tax Rates, 124 Q J Econ 969, 980–81 (2009) (finding that drivers responded less strongly to highway tolls that were automatically deducted from the driver’s account than those that were manually paid to a toll operator by the driver).
• 110See Gamage, 41 Fla St U L Rev at 183 (cited in note 80) (“With regard to market salience, studies of the spotlighting hypotheses found that taxpayers often discount taxes that are not assessed until after market decisions are made.”); Schenk, 28 Yale J Reg at 271 (cited in note 11); Chetty, Looney, and Kroft, 99 Am Econ Rev at 1165 (cited in note 12).
• 111See Gamage and Shanske, 65 Tax L Rev at 27 (cited in note 11) (describing the basic phenomenon of “spotlighting”—when “taxpayers focus[ ] only on certain components of an aggregate price and thereby underestimat[e] the aggregate price”); Krishna and Slemrod, 10 Intl Tax & Pub Fin at 192–93 (cited in note 87). For a review of research regarding the salience of partitioned pricing, see generally Eric A. Greenleaf, et al, The Price Does Not Include Additional Taxes, Fees, and Surcharges: A Review of Research on Partitioned Pricing, 26 J Consumer Psychology 105 (2016).
• 112Schenk, 28 Yale J Reg at 273 (cited in note 11).
• 113Chetty, Looney, and Kroft, 99 Am Econ Rev at 1150–51 (cited in note 12).
• 114See Nussim, 1 Colum J Tax L at 231–33 (cited in note 76).
• 115Chetty, Looney, and Kroft, 99 Am Econ Rev at 1165 (cited in note 12). See also Gamage, 41 Fla St U L Rev at 193–94 (cited in note 80); Sausgruber and Tyran, Tax Salience at *5 (cited in note 59).
• 116Greenleaf, et al, 26 J Consumer Psychology at 116, 119 (cited in note 111); Sean D. Campbell and Steven A. Sharpe, Anchoring Bias in Consensus Forecasts and Its Effect on Market Prices, 44 J Fin & Quant Analysis 369, 388–89 (2009); Nicholas Epley and Thomas Gilovich, The Anchoring-and-Adjustment Heuristic: Why the Adjustments Are Insufficient, 17 Psychological Sci 311, 316–17 (2006); J.D. Trout, Paternalism and Cognitive Bias, 24 L & Phil 393, 406–07 (2005); Korobkin and Ulen, 88 Cal L Rev at 1100–02 (cited in note 61).
• 117See Korobkin and Ulen, 88 Cal L Rev at 1100–02 (cited in note 61); Sunstein, 64 U Chi L Rev at 1188 (cited in note 60); McCaffery, 41 UCLA L Rev at 1916 (cited in note 6).
• 118See Amos Tversky and Daniel Kahneman, Judgment under Uncertainty: Heuristics and Biases, 185 Sci 1124, 1128 (1974).
• 119Id.
• 120See Gamage, 41 Fla St U L Rev at 189–90 (cited in note 80) (discussing the potential effects of the anchoring bias in the tax salience context).
• 121See, for example, Nussim, 1 Colum J Tax L at 229 (cited in note 76); Manju Puri and David T. Robinson, Optimism and Economic Choice, 86 J Fin Econ 71, 75–76 (2007); Oren Bar-Gill, Seduction by Plastic, 98 Nw U L Rev 1373, 1375–76 (2004); Christine Jolls, Behavioral Economics Analysis of Redistributive Legal Rules, 51 Vand L Rev 1653, 1659–62 (1998); Sunstein, 64 U Chi L Rev at 1182–84 (cited in note 60); Neil D. Weinstein, Unrealistic Optimism about Future Life Events, 39 J Personality & Soc Psychology 806, 818–19 (1980).
• 122Howard Latin, “Good” Warnings, Bad Products, and Cognitive Limitations, 41 UCLA L Rev 1193, 1243–44 (1994).
• 123See Schenk, 28 Yale J Reg at 270–71 (cited in note 11). Compare Bar-Gill, 98 Nw U L Rev at 1375–76 (cited in note 121) (discussing an “underestimation bias” in the context of future borrowing that results in part from the optimism bias), with Puri and Robinson, 86 J Fin Econ at 92 (cited in note 121) (arguing that moderate optimism can lead to prudent economic decision-making).
• 124See, for example, Marta P. Coelho, Unrealistic Optimism: Still a Neglected Trait, 25 J Bus & Psychology 397, 404 (2010).
• 125See generally, for example, Chetty, Looney, and Kroft, 99 Am Econ Rev 1145 (cited in note 12).
• 126See text accompanying notes 112–14.
• 127See Gamage, 41 Fla St U L Rev at 195–98 (cited in note 80); Faulhaber, 92 BU L Rev at 1323–25 (cited in note 13); Goldin, 122 Yale L J at 283–85 (cited in note 105); Sausgruber and Tyran, Tax Salience at *23 (cited in note 59).
• 128For anecdotal evidence of websites providing the opportunity to inform consumers about taxes, see Apple iPhone 7 - 32 GB - Black - Unlocked - CDMA/GSM (Google Shopping), archived at http://perma.cc/27FM-2DYJ. See also Anderson, et al, 47 J Mktg Rsrch at 239 (cited in note 35) (concluding that consumers do not react to sales taxes in the context of catalog orders as robustly as they do in the context of Internet orders at least partially because consumers can more easily compare prices on the Internet).
• 129See note 106.
• 130Faulhaber, 92 BU L Rev at 1309–10 (cited in note 13).
• 131See Part II.A.
• 132See note 58 and accompanying text.
• 133Chetty, Looney, and Kroft, 99 Am Econ Rev at 1164 (cited in note 12).
• 134See Part IV.
• 135See notes 107–24 and accompanying text.
• 136See text accompanying notes 127–28.
• 137See text accompanying note 115.
• 138See text accompanying note 127.
• 139See Gamage and Shanske, 65 Tax L Rev at 50 (cited in note 11); Schenk, 28 Yale J Reg at 298 (cited in note 11); Sussman and Olivola, 48 J Mktg Rsrch at S99 (cited in note 15); David J. Hardisty, Eric J. Johnson, and Elke U. Weber, A Dirty Word or a Dirty World? Attribute Framing, Political Affiliation, and Query Theory, 21 Psychological Sci 86, 88 (2010); McCaffery and Baron, 12 Psychology, Pub Pol & L at 117–19 (cited in note 86); McCaffery and Baron, 52 UCLA L Rev at 1759–61 (cited in note 89); Edward J. McCaffery and Jonathan Baron, Heuristics and Biases in Thinking about Tax, 96 Ann Conf Taxn 434, 437–39 (2003). See also generally Rosenberg, 16 Va Tax Rev 155 (cited in note 96) (detailing how people dislike taxes); Christopher C. Fennell and Lee Anne Fennell, Fear and Greed in Tax Policy: A Qualitative Research Agenda, 13 Wash U J L & Pol 75 (2003) (proposing a research agenda regarding tax aversion).
• 140See Sussman and Olivola, 48 J Mktg Rsrch at S99 (cited in note 15) (“Across five experiments, we have demonstrated that people exhibit tax aversion, defined as a tendency to avoid taxes more than other equivalent (or even larger) costs.”); Hardisty, Johnson, and Weber, 21 Psychological Sci at 88 (cited in note 139) (“[P]articipants were more likely to prefer the more expensive product and were more supportive of regulation when the cost increase was described as a carbon offset than when it was described as a carbon tax.”); Gamage and Shanske, 65 Tax L Rev at 50 (cited in note 11) (“[T]he tax-label aversion hypothesis is based on the notion that the mere labeling of a policy as a ‘tax’ can reduce voter support for the policy.”); McCaffery and Baron, 12 Psychology, Pub Pol & L at 119 (cited in note 86) (“In sum, labels matter, and tax tends to be a negative one.”); Homonoff, Can Small Incentives Have Large Effects at *4 (cited in note 96) (“[R]ecent evidence suggests that customers are more likely to avoid any charge that is framed as a tax (as opposed to a fee).”).
• 141Sussman and Olivola, 48 J Mktg Rsrch at S99 (cited in note 15).
• 142Fennell and Fennell, 13 Wash U J L & Pol at 80–81 (cited in note 139). Sales tax holidays are limited periods of time during which a state exempts certain purchases—usually purchases related to children’s education—from sales taxes. Overconsumption during a sales tax holiday thus indicates that people overreact to taxes when they are imposed.
• 143See Sussman and Olivola, 48 J Mktg Rsrch at S100 (cited in note 15). Gamage notes that people are typically averse to being manipulated, and this aversion can affect how firms interact with their customers. See Gamage, 41 Fla St U L Rev at 199–202 (cited in note 80). There appears to be little risk of causing consumers to feel manipulated by the mere advertising of sales-tax-free shopping, though if pretax price raising also occurs, the aversion to being manipulated could reinforce the fairness concerns discussed in Part II.A.
• 144See, for example, Ordering Information (Wayfair LLC), archived at http://perma.cc/SZ4P-X7FJ (“One of the best things about buying through Wayfair is that we do not have to charge sales tax, with a few notable exceptions.”). Indeed, as noted earlier, this existing advantage of remote vendors like Amazon may be a primary driver in seeking out customer-based incentives. See note 44. Essentially, the remote vendor is seeking to maintain the status quo—in which it may take advantage of consumer behaviors such as tax-label aversion to increase demand—when it agrees to invest in the state.
• 145See note 121.
• 146See Faulhaber, 92 BU L Rev at 1336–39, 1342 (cited in note 13); Galle, 67 Tax L Rev at 92 (cited in note 76) (arguing that firms would be incentivized to mask aspects of tax penalties that might affect consumer behavior in a way that is detrimental to the firms); Schenk, 28 Yale J Reg at 267 (cited in note 11) (“The way the tax is framed may make it less prominent.”); Slemrod, 66 FinanzArchiv/Pub Fin Analysis at 21 (cited in note 88).
• 147See Greenleaf, et al, 26 J Consumer Psychology at 116 (cited in note 111); Campbell and Sharpe, 44 J Fin & Quant Analysis at 388–89 (cited in note 111); Epley and Gilovich, 17 Psychological Sci at 311–12 (cited in note 116); Trout, 24 L & Phil at 406–07 (cited in note 116); Korobkin and Ulen, 88 Cal L Rev at 1100–02 (cited in note 61).
• 148See Einav, et al, 104 Am Econ Rev at 3–4 (cited in note 35) (estimating that a sales tax of 10 percent reduces demand by 15 percent, demonstrating a degree of hypersalience); McCaffery, 41 UCLA L Rev at 1907–08 (cited in note 6) (predicting that taxpayers’ biases will give nontaxable fringe benefits “added luster” because of the nontaxable label).
• 149Faulhaber, 92 BU L Rev at 1336–39 (cited in note 13).
• 150Galle, 67 Tax L Rev at 65 (cited in note 76) (“Third parties, especially those who might capture some of the benefit of the price instrument, can be another source of hypersalience.”); Faulhaber, 92 BU L Rev at 1328–30 (cited in note 13).
• 151See note 144.
• 152See note 94.
• 153See notes 109–11 and accompanying text. This analysis assumes that the managers of the firm share quality information with each other, which may be counterfactual for many firms.
• 154See text accompanying note 133.
• 155The tax relief from the two forms of incentives might be expected to have different levels of practical salience to firm owners, provided that the incentives are reported to owners differently. See note 94.
• 156This discussion assumes that the state has decided to provide targeted economic development incentives in the first place; thus it does not enter into the lively debate over whether such incentives should be offered at all. See Oesterle, 6 Ohio St Entrepreneurial Bus L J at 495–98 (cited in note 27) (detailing skepticism regarding the effectiveness of economic development incentives); Edward A. Zelinsky, Tax Incentives for Economic Development: Personal (and Pessimistic) Reflections, 58 Case W Reserve L Rev 1145, 1148–50 (2008) (“[T]he tax incentives states and localities grant are, as a substantive matter, often inefficient and typically unfair.”); Rogers, 4 Georgetown J L & Pub Pol at 101 (cited in note 29) (“State tax incentives do not affect where businesses locate. There is near consensus in the empirical literature on this point.”); Steven M. Rauser, Clearing the Hurdles: Are State Tax Incentives Worth the Effort?, 8 State & Local Tax Law 105, 105–14 (2003); Petrov, Note, 33 Case W Reserve J Intl L at 77–80 (cited in note 26) (documenting the debate over the effectiveness of economic development incentives); Pischak, 8 J State Taxn at 197–203 (cited in note 7).
• 157These issues should affect voters’, and thus the state’s, preferences. Initially, because the different forms of targeted economic development incentives have different levels of practical salience for consumers, people can be expected to have different preferences regarding their use. Tax-averse people may prefer taxes that are less practically salient because such taxes may be less painful to experience, and those same people may prefer the opposite for tax relief. See Schenk, 28 Yale J Reg at 284 (cited in note 11); McCaffery and Baron, 52 UCLA L Rev at 1762 (cited in note 89). Others may desire more-salient tax provisions to ensure the public is able to meaningfully check the growth of government and acts of self-serving politicians. See Galle, 87 Wash U L Rev at 94–98 (cited in note 11); Schenk, 28 Yale J Reg at 287–94 (cited in note 11); Zelinsky, 58 Case W Reserve L Rev at 1146–47 (cited in note 156). Customer-based incentives may be somewhat more transparent to taxpayers than traditional incentives simply because the firm has an incentive to alert consumers in some way to the fact that it is receiving incentives. Though clever firms may find ways to mask the customer-based incentives, a firm receiving traditional incentives would have no apparent incentive to alert taxpayers to that fact. In any event, there is no inherently appropriate level of practical salience for tax provisions; that level is a political matter. See Gamage and Shanske, 65 Tax L Rev at 81–82 (cited in note 11); Schenk, 28 Yale J Reg at 284, 310 (cited in note 11); Sausgruber and Tyran, Tax Salience at *23 (cited in note 59) (observing that the “use of non-salient taxes could allow budget-maximizing politicians and bureaucrats to increase tax revenues beyond what citizens deem appropriate”).
• 158Nussim, 1 Colum J Tax L at 258 (cited in note 76) (describing social welfare as “the aggregate of individual well-being”); Gruber, Public Finance at 50, 52 (cited in note 57). Generally speaking, members of society have heterogeneous preferences—they prefer different things. For example, given $100 to spend on either concert tickets or a new television, Ben might prefer to purchase the tickets while Carly might prefer to purchase the television. The reason for these heterogeneous preferences results from each member of society placing different values on different things. Though the concert tickets and television might both cost $100, Ben personally values the concert tickets at $200 and Carly at $50; the values they each place on the television might be swapped. Therefore, given the choice between paying $100 for one or the other, Ben will buy the concert tickets, and Carly will buy the television. Not only will both Ben and Carly have things that they prefer, both also experience the joy of obtaining those things for less money than they valued them at. Overall social welfare is increased not only by the profit that the sellers generate from selling the tickets and television, but also by the satisfaction Ben and Carly receive. See Gruber, Public Finance at 50, 52 (cited in note 57) (describing social surplus and social welfare).
• 159Galle, 87 Wash U L Rev at 66–67 (cited in note 11); Gruber, Public Finance at 578–80 (cited in note 57). After taxes, consumers are less willing to purchase the same quantity of a good because the price they must pay increases, and vendors are less willing to sell the same quantity of the good as the amount they receive for each sale decreases. Nussim, 1 Colum J Tax L at 234–36 (cited in note 76); Gruber, Public Finance at 578–83 (cited in note 57). For instance, suppose a 5 percent sales tax is levied on all sales and customers bear the full cost of the tax. Taking the example from the prior footnote, the concert tickets and television cost consumers $105 instead of the pretax $100. Ben and Carly still value the tickets and television more than $105 respectively and will still purchase each, but other consumers who value the items at between $100 and $105 will no longer purchase them. This resulting shift in preferences is known as the “substitution effect” of the tax. Nussim, 1 Colum J Tax L at 235 (cited in note 76); Gruber, Public Finance at 36 (cited in note 57). The substitution effect reduces consumer surplus and vendor surplus in an absolute way—some of the pretax social welfare is lost because welfare-generating transactions are not engaged in; this is the deadweight loss due to the tax. Gruber, Public Finance at 51–52, 578–83 (cited in note 57).

  Taxes also generate what is known as the “income effect”: when Ben and Carly make their purchases and bear the burden of the taxes, they each have less income remaining than in the no-tax world (that is, they each are $5 worse off), affecting their remaining preferences. See Nussim, 1 Colum J Tax L at 235 (cited in note 76); Gruber, Public Finance at 36 (cited in note 57). The income effect is often ignored in social welfare analysis because the form of tax does not affect the wealth loss from the tax—the loss is inevitable—and the individual’s loss is offset by the public’s gain. See Gamage and Shanske, 65 Tax L Rev at 62–63 (cited in note 11); Nussim, 1 Colum J Tax L at 235 (cited in note 76). Thus, this discussion largely does not address the income effect. This is not to say that the income effect is not important; it may be. See Galle, 67 Tax L Rev at 66–67 (cited in note 76); Gamage and Shanske, 65 Tax L Rev at 65–69 (cited in note 11).

• 160Gruber, Public Finance at 523 (cited in note 57). As an example, take income as a proxy for social welfare. In a regressive distribution, income is redistributed from low- to high-income people; as a result of the distribution, high-income people end up with a greater share of total income than they had before the distribution. Id. In a proportional distribution, income is not redistributed—each person receives a share of the total income proportionate to their predistribution share. Id. Finally, in a progressive distribution, income is redistributed from high-income people to low-income people; after the distribution, low-income people have a higher share of total income than they had before the distribution. Id.

  The federal income tax is an example of a progressive tax. See IRC § 1(a)–(e) (containing progressive tax rates: as a taxpayer’s income increases, so does her tax rate). On the other hand, state sales taxes are typically regarded as regressive taxes. See Pomp, State & Local Taxation at 6-15 (cited in note 31). A true flat-rate tax would represent a proportional tax. See Joseph Bankman and Thomas Griffith, Social Welfare and the Rate Structure: A New Look at Progressive Taxation, 75 Cal L Rev 1905, 1912 (1987). On pure equity grounds, regressive distributions are typically disfavored; proportional and progressive distributions each have their advocates. For the iconic article discussing the cases for both proportional and progressive taxation, see generally Walter J. Blum and Harry Kalven Jr, The Uneasy Case for Progressive Taxation, 19 U Chi L Rev 417 (1952). For another distinguished foray into this topic, see generally Bankman and Griffith, 75 Cal L Rev 1905 (cited in note 160).

• 161See Nussim, 1 Colum J Tax L at 236 (cited in note 76).
• 162This assumption allows the analysis to focus solely on the different effects on social welfare of the two forms of economic development incentives outside of the state’s economic gains from a business, infrastructure, and employment perspective.
• 163This corollary may prove unrealistic in practice. See note 162.
• 164Not all taxes must generate deadweight loss, but those that do not—such as head taxes or poll taxes—are generally rejected on equity grounds. See Bankman and Griffith, 75 Cal L Rev at 1913, 1920, 1966 (cited in note 160).
• 165For purposes of simplicity, it is assumed that the incentives considered relieve recipients of all of the tax that the incentives target. For instance, if considering the customer-based incentives provided to Amazon, this Section would assume that those incentives relieve customers of all of the sales taxes on their Amazon purchases, not just a portion of those taxes.
• 166See note 58 and accompanying text.
• 167To see this result graphically, return to Figure 1. Assuming the incentives provide full relief of the taxes imposed, then the tax revenue rectangle represents the tax relief the incentives offer. If the vendor keeps prices at the pre-incentives level (P2 in Figure 1), then it will retain all of the tax relief, and the tax revenue rectangle will be converted to vendor surplus.
• 168Gruber, Public Finance at 45–46, 554–55 (cited in note 57).
• 169See Galle, 65 NYU Ann Surv Am L at 540 (cited in note 94) (“Typically, incidence depends on the elasticities of supply and demand for the firm’s products and inputs.”); Chetty, Looney, and Kroft, 99 Am Econ Rev at 1169 (cited in note 12) (discussing the effects of elasticity for the burden of taxes); Slemrod, 61 Natl Tax J at 253 (cited in note 35) (describing how relative elasticities affect who bears the burden of taxes on labor); Gruber, Public Finance at 552–56 (cited in note 57).
• 170See note 167.
• 171Consumers and vendors will likely bear different tax burdens depending on the taxes under consideration. For example, property tax relief may be borne more heavily by the firm than sales taxes (that is, the property tax burden may not be passed along to consumers to the same degree that the sales tax burden is). Thus, the tax a targeted economic development incentive relieves may influence whether it is in the firm’s interest to maintain after-tax prices or to return to pretax prices.
• 172This is not to say that the firm might not later decide to share that vendor surplus with its customers by lowering prices.
• 173See Galle, 65 NYU Ann Surv Am L at 540 (cited in note 54) (“When a firm is liable for a judgment, the economic burden of that judgment is ultimately passed on to real people, whether they are the firm’s owners, its workers, its customers, or even investors in other businesses.”); Rogers, 4 Georgetown J L & Pub Pol at 109 (cited in note 29) (“[T]ax abatements shift money from governments to businesses, and that money is then passed along to consumers, workers and capital owners in the form of wages, lower prices, or higher capital returns.”). See also Gruber, Public Finance at 700 (cited in note 57); McCaffery, 41 UCLA L Rev at 1883 (cited in note 6).
• 174See note 59.
• 175See Parts II.A–B.
• 176See note 60. See also text accompanying notes 98–100.
• 177See Parts II.A–B.
• 178Faulhaber, 92 BU L Rev at 1309 (cited in note 13).
• 179See McCaffery, 41 UCLA L Rev at 1884 (cited in note 6) (discussing how the framing of a tax can generate greater benefits, because of taxpayer biases, by arguing that repealing the corporate income tax, which is less conspicuous, and replacing it with an equal-revenue but more conspicuous tax would diminish “subjective utility” because of loss aversion).
• 180Also, there should not be an adverse income effect resulting from this shift in demand because the consumers are “purchasing” something they desire—the relief of taxes—even if that something should not be desirable from an economically rational point of view. Even if there were an adverse income effect from overreacting to the tax relief, if people sacrifice only their lowest-preference transactions as a result of the income effect, then the net effect on social welfare should be positive. The value of the lost consumption opportunities for the individual is less than the value of the gains to society. See Gamage and Shanske, 65 Tax L Rev at 66–68 (cited in note 11) (observing that “[w]hen taxes reduce individuals’ budgets, the standard models assume that the individuals optimally allocate their (now smaller) after-tax budgets across goods and time periods” and arguing that it should rarely be the case that the impact of the income effect on social welfare outweighs that of the substitution effect because taxpayers should not be expected to “purchase luxury items before necessities” in most circumstances).
• 181See notes 162–63 and accompanying text.
• 182See note 162 and accompanying text.
• 183See, for example, Daphne A. Kenyon, Adam H. Langley, and Bethany P. Paquin, Getting the Incentives Right on Tax Incentives, 69 State Tax Notes 783, 786–87 (2013); Billy Hamilton, Pitching and Hitting: Can State Tax Incentives Be Improved?, 64 State Tax Notes 915, 919 (2012); Timothy J. Bartik, Eight Issues for Policy toward Economic Development Incentives, 10 The Region 43, 44 (June 1996).
• 184See Pollard, 11 Hastings Bus L J at 21–22, 26–27 (cited in note 24); Peter D. Enrich, Business Tax Incentives: A Status Report, 34 Urban Law 415, 423 (2002); Scott J. Ziance, Making Economic Development Incentives More Efficient, 30 Urban Law 33, 41 (1998).
• 185See Enrich, 34 Urban Law at 424–25 (cited in note 184); Ziance, 30 Urban Law at 44–46 (cited in note 184).
• 186See Ziance, 30 Urban Law at 46 (cited in note 184).
• 187See generally Rubin, 70 NYU L Rev 1277 (cited in note 49) (discussing difficulties enforcing agreements between firms and governments, including tax incentive agreements). See also Ziance, 30 Urban Law at 44 (cited in note 184) (discussing legal challenges to enforcing incentive agreements); Pollard, 11 Hastings Bus L J at 21–22 (cited in note 24) (providing an example of a case dealing with the legal issues arising from the escape problem and subsequent clawback attempts).
• 188See Enrich, 34 Urban Law at 425 (cited in note 184):

  [T]he same political dynamics that have fueled the incentive competition over recent decades are likely to deter enactment and enforcement of [clawback] measures with real teeth, and, in fact, there is a danger that relatively toothless accountability measures will merely provide political cover for continued escalation of the interstate subsidy competition.

  See also Ziance, 30 Urban Law at 44 (cited in note 184) (“The costs of enforcing incentive agreements are high. . . . [C]ities face even higher costs from the fear of being labeled ‘anti-business.’”).

• 189See Slemrod, 66 FinanzArchiv/Pub Fin Analysis at 22–26 (cited in note 88); Marjorie E. Kornhauser, Cognitive Theory and the Delivery of Welfare Benefits, 40 Loyola U Chi L J 253, 273, 287–89 (2009) (discussing the detrimental effects on civic virtue due to perceptions of unfairness in the law); Susan Cleary Morse, Using Salience and Influence to Narrow the Tax Gap, 40 Loyola U Chi L J 483, 486 (2009) (discussing the influence of social norms on noncompliance with laws); Amy Gangl, Procedural Justice Theory and Evaluations of the Lawmaking Process, 25 Polit Behav 119, 119–22 (2003) (providing a survey of research into the effects of procedural justice on perceptions of the legitimacy of laws); Bruno S. Frey, A Constitution for Knaves Crowds Out Civic Virtues, 107 Econ J 1043, 1048 (1997) (demonstrating that designing laws suggesting a distrust of citizens leads to lack of civic virtue).
• 190See Slemrod, 66 FinanzArchiv/Pub Fin Analysis at 23 (cited in note 88); Alex Raskolnikov, Revealing Choices: Using Taxpayer Choice to Target Tax Enforcement, 109 Colum L Rev 689, 697 (2009); Michael Doran, Tax Penalties and Tax Compliance, 46 Harv J Legis 111, 131–32 (2009); Gangl, 25 Polit Behav at 119–22 (cited in note 189); Frey, 107 Econ J at 1052 (cited in note 189); Rosenberg, 16 Va Tax Rev at 202–04 (cited in note 96).
• 191See Slemrod, 66 FinanzArchiv/Pub Fin Analysis at 23 (cited in note 88); Leandra Lederman, The Interplay between Norms and Enforcement in Tax Compliance, 64 Ohio St L J 1453, 1488 (2003); Kornhauser, 40 Loyola U Chi L J at 288–89 (cited in note 189); Morse, 40 Loyola U Chi L J at 486 (cited in note 189); Frey, 107 Econ J at 1044, 1049 (cited in note 189); Rosenberg, 16 Va Tax Rev at 197–98 (cited in note 96). See also generally Benny Geys and Kai A. Konrad, Patriotism and Taxation (Max Planck Institute for Tax Law and Public Finance Working Paper No 2016-11, Nov 2016), archived at http://perma.cc/BC9G-J85D (exploring the effects of patriotism on taxpayers’ willingness to comply with tax laws).
• 192See, for example, Bill Kidd, Lawmakers Considering Sales Tax Break for Amazon, 60 State Tax Notes 929, 929 (2011) (quoting Eric Bearse, spokesman for the Alliance for Main Street Fairness, as saying, in an effort to urge lawmakers to reject targeted tax incentives for Amazon, that Amazon’s attempt to “get special treatment from Texas lawmakers should have every business owner and taxpayer outraged. . . . Texans will not stand by and let the government give preferential treatment to one out­of­state company that gives it an advantage over existing Texas businesses”); John Buhl, California Democrats Speak Out against Proposed Amazon Tax Deal, 2011 State Tax Today 173­2 (Sept 7, 2011) (quoting California Assembly Speaker John Perez as stating, in opposition to targeted tax incentives for Amazon, “This is a matter of simple fairness. . . . Every brick and mortar store in California must collect and pay sales taxes. That means thousands of small businesses, as well as large retailers, are shouldering a burden that online retailers like Amazon have been exempting themselves from, and that’s not fair”).
• 193See Korobkin and Ulen, 88 Cal L Rev at 1092 (cited in note 61) (observing that people are more optimistic about events if those events are perceived to be within their control); Jolls, 51 Vand L Rev at 1675 (cited in note 121).
• 194For a discussion of the difference between nontaxation and nonenforcement in the context of actions by the Internal Revenue Service, see generally Lawrence Zelenak, Custom and the Rule of Law in the Administration of the Income Tax, 62 Duke L J 829 (2012).
• 195See Thimmesch, 118 W Va L Rev at 177–81 (cited in note 42) (discussing effects on civic virtue of the nonenforcement of use taxes); Leigh Osofsky, The Case for Categorical Nonenforcement, 69 Tax L Rev 73, 124–26 (2015) (observing that categorical nonenforcement might lead to less compliance with the law while also providing reasons why such nonenforcement might not have that effect); Zelenak, 62 Duke L J at 854 (cited in note 194) (“It is a commonplace that nonenforcement (or severe underenforcement) of one law may breed among the citizenry a general disrespect for the law.”). See also Michael Sant’Ambrogio, The Extra-legislative Veto, 102 Georgetown L J 351, 387–93 (2014) (discussing the harms resulting from presidential actions designed to not enforce enacted laws).

  Professor Leigh Osofsky makes a compelling case that explicit categorical nonenforcement by an agency can be more legitimate than other forms of nonenforcement in the real world when administrative resources are not sufficient to fully enforce the law. Osofsky, 69 Tax L Rev at 131–32 (cited in note 195). However, the type of nonenforcement involved in the provision of customer-based incentives is different; the nonenforcement is not the result of insufficient resources (though it may certainly help relieve pressure on limited resources), but it is the result of a state’s economic development policy. This difference may result in more erosion of civic virtue than the “limited resources” approach to nonenforcement, as Osofsky observes while taking issue with the line between the two forms of nonenforcement for purposes of constitutional analysis. Id at 115–24 (addressing concerns about categorical nonenforcement arising from policy-based decisions and the effect of categorical nonenforcement on the rule of law).

• 196See Allison Christians, Avoidance, Evasion, and Taxpayer Morality, 44 Wash U J L & Pol 39, 47 (2014); Zelinsky, 58 Case W Reserve L Rev at 1150–51 (cited in note 156) (suggesting that nonfavored firms may retaliate to tax incentives by “demand[ing] the same relief granted to others” or “threatening to depart”); Smith, 17 J State Taxn at 4 (cited in note 7) (“[T]ax incentives . . . are inequitable because they favor manufacturing and mobile businesses that can threaten to relocate.”).
• 197See notes 43–46 and accompanying text.
• 198But see Thimmesch, 118 W Va L Rev at 160–83 (cited in note 42) (discussing the various long-term harms that might result from the nonenforcement of use taxes).
• 199See id at 158–60 (detailing states’ efforts to enforce use tax compliance). See also note 45.
• 200See Osofsky, 69 Tax L Rev at 97–98 (cited in note 195) (observing how the public exposure of categorical nonenforcement can result in action against such nonenforcement, using the example of low partnership audit rates by the Internal Revenue Service, or may increase the legitimacy of the nonenforcement by spurring public debate over its merits).
• 201See text accompanying note 29.
• 202See text accompanying note 26.
• 203See Schaefer, 28 NM L Rev at 309–10 (cited in note 27); Rauser, 8 State & Local Tax Law at 113 (cited in note 156); Key and Smith, 15 J State Taxn at 6 (cited in note 27); Benedicte E.F. Mathijsen, Enterprise Zones as Tools of Urban Industrial Policy, 6 Mich Yearbook Intl Legal Stud 233, 249 (1984).
• 204See Schenk, 28 Yale J Reg at 284 (cited in note 11); McCaffery and Baron, 52 UCLA L Rev at 1762 (cited in note 89).