The hidden costs of cliff effects in the Internal Revenue Code.

• Author: Viswanathan, Manoj
• Position: II. Assessing the Burden of Cliff Effects D. Equity Concerns of Income-Based Cliff Effects through Conclusion, with footnotes, p. 955-981

4. Equity Concerns of Income-Based Cliff Effects

   Horizontal equity demands that tax provisions not treat similarly situated taxpayers differently. (114) Because, tautologically, two identically situated taxpayers cannot be treated differently, (115) a tax provision's compliance with horizontal equity requires defining what it means for taxpayers to be "similarly situated." This is done by reference to some metric by which taxpayers can be compared. Most commonly, this metric is income. Consider two taxpayers who earn identical income but from different sources. Horizontal equity requires that the taxpayers have identical income tax liabilities, assuming no governmental interest exists in promoting one income source over another. (116) Vertical equity requires that, given two otherwise identical taxpayers, a taxpayer with more income should not pay less in taxes. (117)

   For cliff effects where the reference metric is something other than income, the line drawn might properly divide taxpayers into proper benefit-receiving and nonbenefit-receiving groups. For example, a transportation subsidy for taxpayers younger than sixteen years old might be appropriate if these taxpayers are not permitted to obtain driving licenses before their sixteenth birthday. A sixteen-year-old taxpayer would experience a cliff effect on her sixteenth birthday but also become eligible for a driving license.

   Cliff effects based on income, however, necessarily violate tenets of both horizontal and vertical equity: two nearly identically situated taxpayers can, by virtue of slight differences in income, have significantly different tax liabilities. To satisfy horizontal equity, the differences between taxpayers subjected to the cliff effect and those not subjected to the cliff effect must render these taxpayers significantly dissimilar. Although cliff effects are used to means-test tax provisions conferring benefits, there is no meaningful distinction between taxpayers just next to either side of the income threshold of the cliff effect.

   Income-based cliff effects also contravene vertical equity. The proposition that a higher-earning taxpayer should pay no less in taxes than an otherwise identical but lower-earning taxpayer implies that a higher-earning taxpayer should not, by virtue of the Internal Revenue Code, be left in a worse economic position post-tax than the lower-earning taxpayer. If a taxpayer's income qualifies her for a tax benefit, her net economic position improves to some minimum standard. If an otherwise identical taxpayer with more income pre-tax is ineligible for the benefit by virtue of her additional income and is in a worse economic position post-tax than the lower-earning taxpayer, this minimum standard is not satisfied and, on a post-tax basis, vertical equity is violated. As demonstrated above, at the income threshold at which the cliff effect is triggered, taxpayers suffering a cliff effect are in a worse economic position relative to a lower-earning taxpayer whose income is just short of the cliff effect threshold. Although a given taxpayer subjected to the cliff effect might earn enough to compensate for the economic harm imposed by the cliff, there exists a range over which taxpayers are worse off for having earned the additional income. The costs, explicit and implicit, required to earn the additional income only exacerbate the consequences of income-based cliff effects. (118)

   For every violation of equity, whether horizontal or vertical, a theoretical minimum dollar amount exists that can be transferred to the suffering taxpayer to cure the equity violation. This "equity cost" represents the cost of modifying a tax provision that is structurally unsound on equity grounds to a provision that is not. The term "equity cost," as used in this Article, is an aggregate microeconomic metric that represents the net economic loss suffered by all taxpayers who are in a worse economic situation post-tax than they would have been had they not exceeded the cliff effect threshold. If the cliff effect creating the equity cost is an income-based cliff effect attached to a means-tested tax provision, the equity cost represents a flaw in the implementation of the tax provision. If the tax provision is intended to benefit a group of taxpayers who are means-tested on a pre-tax basis by increasing their economic position, the tax provision should not make these beneficiaries better off than a group of taxpayers ineligible for the benefit by virtue of earning more. Either the subsidy provided by the tax provision is being awarded to taxpayers who do not need it, or the subsidy is not being provided to those taxpayers who do.

   In addition to benefitting some taxpayers with limited means, a tax provision with an equity cost is penalizing some taxpayers for earning more. Not every taxpayer with an income beyond the cliff effect threshold suffers an equity loss; at some level of income greater than the cliff effect threshold the economic loss of the equity cost of the cliff effect is outweighed by the additional income. This equity cost is borne by those taxpayers who, in retrospect, would have been better off economically had they earned less income. Thus, cliff effects may violate both horizontal and vertical equity by causing taxpayers who are nearly identical in economic status to have dramatically different tax burdens.

5. Efficiency Concerns of Income-Based. Cliff Effects

   Measuring the efficiency of a given tax provision first requires determining the objective (or objectives) of the tax provision. An efficient tax provision will accomplish these objectives at a low cost. The primary objective of the tax on income, for example, is raising revenue. (119) One measure of the cost of a tax provision is the extent to which the tax provision interferes with behavior that would have occurred in the absence of the provision. (120) For example, if a taxpayer would work for no less than $12 per hour and is in a 40% marginal tax bracket, her pre-tax wage must equal twenty dollars per hour. (121) Pre-tax, a wage of twelve dollars per hour is sufficient to incentivize the taxpayer to work. Post-tax, a wage between twelve dollars and twenty dollars per hour will not be sufficient, representing the cost of the 40% marginal tax rate. Over this range of offered wages, the tax system has changed the taxpayers behavior and created inefficiencies by preventing behavior that both employee and employer find economically advantageous pre-tax. (122)

   For tax provisions that are not intended to change behavior, the classic measure of efficiency (or lack thereof) is the "deadweight loss," or "excess burden" of the provision. (123) The efficiency of these tax provisions is generally measured by examining the loss of surplus to both a consumer and producer when comparing pre- and post-tax behavior. (124) This lost surplus rises with the square of the marginal tax rate. (125) Cliff effects based on income, by definition, impose a marginal tax rate of greater than 100%. The deadweight loss associated with a tax provision with a cliff effect is greater than a tax provision that is phased out at some rate less than 100%. (126) Put differently, any tax provision not intended to change behavior that utilizes a cliff effect based on income to eliminate the benefit can be made more efficient by using a phaseout (at a rate less than 100%) instead of the cliff effect.

   Other tax expenditures are not solely intended to change behavior; rather, they are meant to also confer a benefit on some group of taxpayers in order to accomplish some socially valuable set of goals. For example, the Earned Income Tax Credit, in addition to incentivizing working, is also intended to bring people out of poverty. (127) The child tax credit is also intended to help low-income households (with children) out of poverty but is not intended to incentivize households to have more children. (128) However, regardless of the particular objective of the tax provision in question, efficiency is improved when the objective is achieved with a lower cost.

   Many tax expenditures, however, are intended to change behavior. For these provisions, if the behavior that the tax provision seeks to incentivize would occur in the absence of the subsidy provided by the tax provision, then the tax provision is inefficient--the federal government is essentially paying for something it is already getting for free. (129)

6. Quantifying the Aggregate Cost of Income-Based Cliff Effects

   Several scholars have addressed issues confronted by low- and moderate-income taxpayers subjected to varying marginal tax rates, including those associated with cliff effects based on a taxpayer's income. These scholars typically focus on a small subset of taxpayers, such as those residing within a state or other clearly defined region for which there exists an available data set. (130) Other studies use a single-family composition (i.e., a single parent with two children) and assess the consequences to the family as income increases. (131) These analyses are most important when the effects of direct transfer programs outside of the Internal Revenue Code are considered in conjunction with federal expenditures. (132) The result of these efforts has been to calculate "effective" marginal tax rate profiles for narrow selections of taxpayers. (133) For effective marginal tax rates that rise to the level of a cliff effect, the result is straightforward: taxpayers are incentivized to either earn enough income to overcome the loss created by the cliff effect or reduce their income to not be subjected to the cliff effect. (134) This change in behavior, to the extent it occurs, represents an inefficiency generated by the imposition of the cliff effect: otherwise desirable behavior--i.e., an individual performing socially desirable work--is no longer being performed.

   However, low- to moderate-income taxpayers, as discussed previously...