By discussing only how the candidates would change the tax system, reports suggest that the current tax system is somehow optimal. Yet with the top fifth of all earners currently paying 67 percent of all federal taxes, the baseline tax rates should be reexamined as well.

Resident Scholar Andrew G. Biggs

Tax policy has two main goals: fairness and efficiency. Fairness encompasses philosophical values regarding how the tax burden should be distributed based upon the ability to pay. Efficiency incorporates economic consideration of how incentives built into the tax code affect individuals' willingness to work or save. Deciding whether a given tax plan maximizes fairness and efficiency is a difficult task under the best of circumstances. But given how most information pertaining to the two presidential candidates' tax plans is presented in the press, it is all but impossible to say much about either fairness or efficiency.

The media typically report how each candidate's plan would change the tax code relative to current law: the size of the typical tax cut or increase that accrues to low, middle or high earners. So it is widely reported that John McCain would mostly cut taxes for the rich while Barack Obama would mostly cut taxes for the middle class. But how much a given income group gains or loses from a specific tax reform tells us nothing about either the fairness or efficiency of the tax code after such changes have been implemented.

According to the Tax Policy Center, a joint venture between the Urban Institute and Brookings Institution, around 78 percent of the McCain tax cut would accrue to the top fifth of income earners, with almost 30 percent going to the highest 1 percent. This seems inequitable on its face, a point the Obama campaign and the press focus on.

But can we conclude that the rich would pay too little taxes under the McCain plan? Not really, because most media reports do not reveal the resulting share of the tax burden borne by the highest earners.

As it happens, the top fifth of earners currently pay 67% of all federal taxes--including not just income taxes, but payroll taxes, corporate taxes and death taxes. The top 1% of earners pay 26% of all federal taxes.

If the McCain proposal were passed, the top fifth would actually pay a greater share of total federal taxes and the top 1%'s share would decline by only 0.3%. In other words, high earners carry the vast majority of the federal tax burden and, despite what the media portrays as a shift from Scandinavian egalitarianism to Latin American inequity, would continue to do so under Mr. McCain's plan.

The fact that high earners pay the vast majority of all federal taxes will come as a surprise to most Americans, who believe the middle class bears the tax burden while the rich get off scot free. A 2007 Gallup poll found that 66% of Americans believe upper-income people pay too little federal income taxes. Media reports play into these beliefs by focusing on how the candidates' plans would change the tax code, not on what the tax code would look like after those changes took place.

Moreover, can we say anything about whether the candidates' plans improve economic efficiency? Only if we look at whether the plans reduce marginal tax rates--the amount of additional taxes paid for each additional dollar earned--which determine incentives to work, save and create jobs. But media reports commonly tell us nothing about these issues.

As it happens, the McCain proposal would maintain current income tax rates and lower corporate taxes to help American businesses--which ultimately provide American jobs and pay American wages--compete in a global economy. By contrast, the Obama plan would increase marginal income tax rates on high earners and raise taxes on capital gains and dividends, with the likely effect of contracting labor supply, reducing rewards for saving, and increasing incentives to avoid taxes. All but the very lowest income Americans would face higher marginal tax rates under the Obama plan than the McCain proposal.

The media focus on how the candidates' plans would change the tax code implicitly assumes that the current distribution of tax burdens and marginal tax rates is somehow optimal. But there is no reason to believe this. We can argue whether the Obama plan is fairer or more economically efficient than the McCain plan, but only if we focus on the end product. How the presidential candidates' plans would change the tax code matters far less than where the tax code would stand when those changes are accomplished.

Andrew G. Biggs is a resident scholar at AEI. Kent Smetters is a visiting scholar at AEI.