January 31, 2019

This week’s release of the annual Budget and Economic Outlook by the Congressional Budget Office also included revisions to their official estimates of how much money is raised by existing transportation excise taxes. And those numbers allow us to estimate how much money would be raised if those taxes were increased.

Current yields. In their ten-year forecast, CBO estimates that each penny of existing gasoline and diesel excise tax rates will bring in the following amounts to the Highway Trust Fund, in billions of dollars (“special fuels” mostly means propane):

Because of the increasing fuel efficiency of automobiles, the amount raised per penny of fuel taxes is projected to steadily decrease as the average miles per gallon of the fleet on the road continues to increase.

Demand inelasticity. Using that data to extrapolate how much money would be raised by an increase in those rates requires an assumption of what economists call “demand elasticity” – how much do changes in the price of something affect the amount of that something that is purchased? Gasoline has always been notoriously “inelastic” – variations in the price don’t have much effect on demand, since people and goods usually have to go where they have to go, when they have to go, no matter how much a gallon of fuel costs. Obviously, demand is not completely inelastic – if we immediately slapped a European-style $5.00 per gallon federal tax on gasoline, you can be certain that there would be an immediate and measurable decrease in demand as carpooling jumped and leisure travel slacked off. But the price of gasoline routinely fluctuates by 30-40 cents per gallon over the course of a few months without noticeably affecting demand levels, and all the serious tax proposals we have heard are far less than that, so we are just assuming complete inelasticity of demand in our calculations. (Also, the author doesn’t have the econ skills to model elasticity.)

A nickel per year – U.S. Chamber/Earl Blumenauer proposal. A year ago, the U.S. Chamber of Commerce proposed a 25 cent per gallon motor fuels increase (5 cents per per year for five years), with indexation thereafter for both inflation and the annual increase in fleet fuel efficiency. And earlier this month, Bloomberg’s Shaun Courtney got a copy of a draft bill (dated December 2018) from Rep. Earl Blumenauer (D-OR) that would also provide a 5 x 5 gas/diesel tax increase (indexed thereafter by inflation alone, not fuel efficiency increases), which was a shift from the bill he introduced in 2017 (H.R. 1458, 115th Congress) which would have increased the taxes by 15 cents over 3 years (8-4-3).

Since both the nation’s biggest business lobby and a senior Democrat on the House Ways and Means Committee both support 5 x 5 plus indexation, let’s guesstimate how much money this would raise.

Using the total yield numbers above, and ignoring the fact that the CBO numbers are for October 1 to September 30 fiscal years and the tax increases in the Blumenauer bill take effect on January 1 of each year (so we treat calendar and fiscal years as identical, which isn’t quite right, but there’s not much else we can do), and again assuming complete inelasticity of demand, here is the amount of money that the draft Blumenauer bill would raise for the Highway Trust Fund, in billions of dollars:

(After FY24 we use the annual CPI inflation numbers in the latest CBO economic forecast, which average 2.34% per year.)

Using this methodology, a 25 cent tax increase over 5 years, plus CPI-based indexation thereafter, would bring in an extra $372.5 billion or thereabouts for the Highway Trust Fund over the ten-year budget forecasting window.

Income tax offset. However, an increase in transportation excise taxes does not occur in an economic vacuum. Increasing excise taxes deposited in the Highway Trust Fund would also reduce the amount of income taxes received by the general fund of the Treasury. Imagine that you are Trucking Company X and your fleet currently pays $100 million per year in fuel costs. Congress then increases the motor fuel tax, causing your fuel costs rise to $105 million per year. When you fill out your corporate tax returns, you just increase the amount of your deductible business expenses by $5 million. This causes the amount of your taxable income to decrease by $5 million, and the IRS loses some percentage of that $5 million when you file your income taxes. This exercise is repeated by every business and every individual every time that Congress increases an excise or payroll tax – excise and payroll tax increases always lead to less net taxable income for someone, somewhere, which leads to lower income tax receipts.

It is impossible to compute these amounts precisely, so budget estimators used to use a 25 percent rule of thumb – for every dollar of excise or payroll tax increase, income tax receipts were assumed to drop by 25 percent. The changes in the income tax rates in the 2017 tax law changed all that, so now the Joint Committee on Taxation uses a different estimator, fluctuating each year, that was last updated in March 2018 in this document. (Again we have the calendar year vs fiscal year problem, but we will just push on through.)

Using the JCT numbers and bearing in mind that the CY/FY problem makes these approximate, here is the amount of money the general fund would lose from the excise tax increase described above:

The offset doesn’t matter for the Highway Trust Fund (it is the general fund’s problem), but it does have to be taken into account for purposes of budget enforcement, PAYGO, sequestration, etc.

The 80-20 split. Every gas/diesel tax increase from 1982 onwards that has been deposited in the Highway Trust Fund has been split 80-20 between the Highway Account and the Mass Transit Account. The Blumenauer bill is explicit that this split will continue and, while it was not spelled out in the Chamber’s proposal last year, we were told that the 80-20 split was assumed there as well.

Highway Trust Fund solvency under Chamber/Blumenauer. We split the gross total above 80-20 and plugged it into the new CBO baseline HTF cash flow forecast. Here is what we came up with, in billions of dollars.

Assuming that new spending increases above baseline would start at the end of the FAST Act (i.e. take effect starting in FY 2021), an 80-20 split of the Chamber/Blumenauer tax increase would allow an average of about $20 billion per year in highway spending above baseline ($20/year x 9 years = $180 which still leaves about $5 billion as a cash cushion). This would also allow an average of $3 billion per year in mass transit spending above baseline ($3/year x 9 years – $27 billion which leaves just over $1 billion as a cash cushion).

At this point, readers who can do math in their heads are saying, wait a second, I thought an 80-20 split meant that highway spending increases would be 4 times higher than transit spending increases, but 20 to 3 is more like 6.7 to 1. What gives?

Well, it seems that under the FAST Act and previous authorization laws, the Mass Transit Account of the Highway Trust Fund has been allowed to overspend its dedicated tax revenues by a much higher percentage than the highway program has overspent Highway Account dedicated revenues. To put this in perspective, we asked the question:

What is the minimum tax increase needed for baseline HTF solvency? We kept throwing around gas/diesel tax increase numbers for various tax rates, using the above methodology, at the new CBO baseline for Trust Fund cash flow until we found the bare minimum numbers necessary to maintain solvency for a decade. To save time, we just assumed an immediate increase effective October 1, 2o19 (no multi-year phase-in, no indexation), complete demand inelasticity, and an 80-20 highway transit split. We found that to keep the Highway Account solvent for a decade at baseline spending levels, we would need 80 percent of an immediate 8.5 cent per gallon gas/diesel tax increase:

However, in order to keep the Mass Transit Account solvent, we need 20 percent of an immediate 13.5 cent per gallon tax increase:

It has become obvious that allowing mass transit programs to overspend their dedicated revenues by a greater percentage than highways overspend theirs means that the traditional 80-20 split for new revenues may not work anymore – it will make it more difficult to fund transit if an overall tax increase is small, or it will give highways a significantly larger increase in allowable spending if the tax increase is larger.