Payroll Tax Reduction - One Approach to a Carbon Tax

An impressive new report has been released by the Brookings Institution on the issue of carbon taxes: A Green Employment Tax Swap: Using a Carbon Tax to Finance Payroll Tax Relief by economist Gilbert E. Metcalf. The report is available here (PDF).

The proposed policy is straightforward enough:

To illustrate how a carbon tax could be implemented, this policy brief discusses a Green Employment Tax Swap (GETS) in which a tax of $15 per metric ton of carbon dioxide (CO2) is used to rebate the federal payroll tax on the first $3,660 of earnings per worker.

A CO2 tax of $15 per metric ton (equivalent to a $55 tax per metric ton on carbon) is relatively modest compared to other plans, but would raise nearly $90 billion for the federal government, at 2005 emission rates.

This tax would necessarily cause the cost of many items, particularly fossil fuels, to rise. Metcalf estimates the price rise this level of carbon tax would cause, assuming no behavioural changes and that the entire cost of the tax will be borne by consumers:

Consumer Price Increases from a Carbon Tax

Electricity and Natural Gas (up 14.1%)
Home Heating (up 10.9%)
Gasoline (up 8.8%)
Air Travel (up 2.2%)
Other Commodities (up 0.3% to 1.0%)

In reality, the price increases are likely to be somewhat smaller, since the entire cost of the tax is unlikely to be borne by consumers, but these provide good estimates.

Needless to say, these increases in prices by themselves would cause a financial hit on families. However, families also have more money in their pocket, thanks to the reduction in payroll tax rates. These two effects do not cancel themselves out for each group. Assuming no behavioral changes, Metcalf finds the net effect that families in the lowest 3 deciles and the top decile will pay more money under the new tax regimes, and middle class families will pay less. The net effects are relatively modest, with the largest losses going to the highest income families (the 10th decile) with an average of $164 extra spent a year. The largest gains go to the 7th decile, who now have $151 extra in their pocket under the new regime.

Overall, Metcalf has done what he claims - he has found a tax swap which allows for the implementation of a carbon tax, which is also relatively neutral to all income levels.

In the end, I disagree with the choice of using payroll taxes as the type of tax to reduce. Metcalf's choice for using payroll taxes are straight-forward enough:

In fact, for nearly three-quarters of all households, payroll taxes are their single largest tax to the federal government. Because the payroll tax is a flat-rate tax up to a payroll limit of $90,000 (as of 2005), it is generally acknowledged to be a regressive tax. Using the carbon tax to reduce the payroll tax burden would reduce the tax burden on working households, especially for those households earning less than the median income in the United States.

All of that is true. Unfortunately, there is one point left out of the analysis - that payroll taxes are some of the most economically efficient taxes we have. In the article Canadian Conservatives Promote Cutting Consumption Taxes we discussed the concept of a "Marginal Efficiency Cost" (MEC) which measures the damage is done to the economy for every extra dollar generated in tax revenue . In 1997 the Canadian government estimated the MECs for four classes of taxes and found the following:

MEC in CDN$ for 4 classes of taxes

$1.55 - Corporate Income Taxes
$0.56 - Personal Income Tax
$0.27 - Payroll Tax
$0.17 - Sales Tax

While payroll taxes are regressive, they also cause almost 6 times less damage to the economy than corporate income taxes. This is important, because a carbon tax, similar to every other tax, will act to slow economic growth. By replacing a carbon tax with the most inefficient taxes first, we can ensure that a minimal amount of damage occurs to the economy. In fact, a straight swap of carbon taxes for corporate income taxes is likely to boost economic growth, given how particularly inefficient carbon taxes are.

By swapping the carbon tax with corporate income taxes, we are not switching one regressive tax with another regressive tax. To ensure that lower income families do not get hurt too much by the switch, a carbon tax rebate of some sort would be necessary. The Metcalf report suggests that at $15 per ton of carbon, the lowest decile household would pay $289 a year in carbon taxes. Thus a rebate in that range (if not higher) would be required to ensure that low income families do not take too much of a hit.

Overall, I found the Metcalf report to be an impressive piece of research, even if I believe he has not found the ideal tax swap.