A good economic analysis of the McCain and Clinton proposals must answer two questions:
- What effect does the federal gasoline tax have on price?
- What is the incidence of the tax? That is, how much of the federal gas tax is paid by consumers, and how much is paid by producers?
To answer those questions, we have to delve into some technical economics. We must think about supply and demand curves. But have no fear — this will all make sense.
It’s generally accepted that consumers do not change their use of gasoline over the short term despite large swings in price. That’s why the near-doubling of the price of oil over the past year has resulted in only a tiny decrease in U.S. consumption of oil. Likewise, large swings in price do not change producers’ production of gasoline. That is because it takes many years to establish new oil fields, build transportation systems to move the oil to market and build refineries to process the oil into gasoline.
In economics jargon, both oil supply and demand are “inelastic” — they show only a minimal short-term response to changes in price. As shown in Figure 1, we should think of both the supply and demand curves as being very steep.
If you remember your ECON 101, you know that the market price and quantity of a good are established by the intersection of the supply and demand curves. These levels are often denoted as P* and Q*.
But what if gasoline is taxed? You might think that the consumer pays all of the tax through a higher price on the good, or you might think that the producer pays all of the tax through reduced profits. In fact, it is usually the case that both producers and consumers shoulder some of the tax burden. To determine who shoulders the majority of the burden, we have to look at the shape of the supply and demand curves.
Think of the tax as a rectangle that stretches from the Y‑axis of our diagram (that is, the axis that shows price) to the intersection of the curve. The height of the rectangle is the amount of the tax — in the case of the federal gas tax, that’s 18.4 cents. Economic forces will push the rectangle to nestle as close to the intersection as it can get without either of the corners of the rectangle pushing past one of the curves. We show this in Figure 2.
The portion of the tax rectangle that lies above the horizontal line denoting P* represents the amount of the tax that falls on consumers; the amount below the P* line represents the amount that falls on producers. Consumers ultimately end up paying price P+ while producers receive P-. The actual quantity of the good sold will be Q’.
Notice that, if the tax in Figure 2 were to be repealed, the price to consumers would fall from P+ to P*, and the revenue per unit received by producers would increase from P- to P*. The total quantity of the good sold would increase from Q’ to Q*
If either the supply or demand curves were to have a different slope, it would change the incidence of taxation. If the demand curve were flatter, the tax rectangle would move downward so that most of the tax would fall below the P* line, meaning that producers would pay more of the tax. Conversely, if the supply curve were flatter, the rectangle would move upward and consumers would pay more of the tax.
Many of the critics of a gas-tax holiday focus on the short-run inelasticity of supply in the summer and argue that the federal gas tax now falls largely on producers. They claim that a gas tax holiday would bring little benefit to consumers but big profits for producers.
We’re not convinced by their argument, for a few reasons:
First, the critics seem to assume that the supply of oil is not just inelastic, but perfectly so — that is, that the supply line is completely vertical and no increase in price will induce more supply in the short term. For instance, New York Times columnist Paul Krugman wrote in a 2000 column, “The quantity of oil available for U.S. consumption over the near future is pretty much a fixed number: the inventories on hand plus the supplies already en route from the Middle East.” But the oil supply is not perfectly inelastic and it could become more elastic if inventory-holders (including the U.S. government and its Strategic Petroleum Reserve) decide to tap into their stocks. Though oil production may not be able to increase quickly, inventories can be emptied quickly when gas prices reach an attractive-enough price relative to expectation of future prices.
Second, though the short-run world supply of gasoline may be fixed, the supply to the United States is not. We can induce gasoline imports by lowering our taxes on gasoline relative to taxes elsewhere, allowing producers to gain back some of the profits they lose on the tax. Taxes in Europe are already very high, especially on diesel. This difference is enough to send European gasoline into the U.S. East Coast in the summer. Lowering our tax makes sending supply to the United States even more attractive to producers.
Finally, and perhaps most importantly, the critics’ argument gives no attention at all to the demand curve, yet the demand curve is just as important in determining the incidence of taxation as is the supply curve. Most analysts assume oil and gas demand is highly inelastic. In fact, recent research suggests that demand for gasoline has grown increasingly inelastic over the last few decades. If supply is more elastic than conventional wisdom suggests and demand is more inelastic, then more of the tax falls on consumers and a gas tax holiday could bring price relief to consumers. (But remember: the proposal is a temporary gas-tax holiday during the summer. Summer demand may be more elastic than demand during other times of the year because the extra summer travel is discretionary rather than work- or school-related.)
While the McCain and Clinton plans are being debated in Washington, some states are also considering implementing holidays for their state gas taxes. Those proposals could bring significantly more price relief to consumers than a federal holiday. This is because state supplies of gasoline are elastic. Gasoline can be rerouted easily from one state market to another depending on consumer demand. Because of this, a large portion (if not most) of the state gas tax falls on consumers — or at least on consumers who live far enough from the state line that they can’t buy gas routinely in lower-tax states.
One common criticism thrown at gas tax holiday proposals is that a “tax break would prompt a jump in demand that would push up prices.” But, as we’ve noted before, and as most critics agree, gasoline demand is believed to be highly inelastic. The critics cannot logically claim both that demand for gas is inelastic and also that the quantity demanded will increase dramatically if the price falls from its current level.
It is plausible that short-run supply is more elastic than many of these critics presume. Thus, a temporary suspension of the gas tax suspension could bring consumers some price relief, depending on the elasticities of supply and demand. And even if supply is more inelastic, consumers would still benefit from an increase in supply that would result from the gas tax holiday.
So is suspending the tax a good idea? What we have tried to show is that consumers probably shoulder some of the gas tax burden and thus would experience some savings if the tax were suspended. But at the same time, suspending the tax would reduce revenues that go to pay for road and other transportation projects. Would the benefit of suspending the tax offset the cost of lost funding?
Elsewhere, one of us has argued against suspending the tax because it would hide the true cost of driving from motorists. The other has criticized the gas tax itself because there are better, more direct ways to charge motorists for highway use and pollution.
More generally, even if economics can clearly identify the costs and benefits and the identities of the winners and losers, economists have no particular qualification to proclaim that a particular distributional outcome is preferred. To come to such a conclusion, values outside economics must be introduced into the analysis.
