To understand the economics of taxes, it’s helpful to look at a simple type of tax known as an excise tax—a tax charged on each unit of a good or service that is sold. Most tax revenue in Canada comes from other kinds of taxes, which we’ll describe later in this chapter. But excise taxes are common. For example, there are excise taxes on gasoline, cigarettes, and fuel-inefficient vehicles or ones with air conditioning. Many local governments impose excise taxes on services such as hotel room rentals. The lessons we’ll learn from studying excise taxes apply to other, more complex taxes as well.

Suppose that the supply and demand for hotel rooms in the city of Potterville are as shown in Figure 7-1. We’ll make the simplifying assumption that all hotel rooms are the same. In the absence of taxes, the equilibrium price of a room is $80 per night and the equilibrium quantity of hotel rooms rented is 10 000 per night.

Now suppose that Potterville’s government imposes an excise tax of $40 per night on hotel rooms—that is, every time a room is rented for the night, the owner of the hotel must pay the city $40. For example, if a customer pays $80, $40 is collected as a tax, leaving the hotel owner with only $40. As a result, hotel owners are less willing to supply rooms at any given price.

What does this imply about the supply curve for hotel rooms in Potterville? To answer this question, we must compare the incentives of hotel owners pre-tax (before the tax is levied) to their incentives post-tax (after the tax is levied).

From Figure 7-1 we know that pre-tax, hotel owners are willing to supply 5000 rooms per night at a price of $60 per room. But after the $40 tax per room is levied, they are willing to supply the same amount, 5000 rooms, only if they receive $100 per room—$60 for themselves plus $40 paid to the city as tax. In other words, in order for hotel owners to be willing to supply the same quantity post-tax as they would have pre-tax, they must receive an additional $40 per room, the amount of the tax. This implies that the post-tax supply curve shifts up by the amount of the tax compared to the pre-tax supply curve. At every quantity supplied, the supply price—the price that producers must receive to produce a given quantity—has increased by $40.

The upward shift of the supply curve caused by the tax is shown in Figure 7-2, where T is the amount of tax, S₁ is the pre-tax supply curve, and S₂ = S₁ + T is the post-tax supply curve. As you can see, the market equilibrium moves from E, at the equilibrium price of P_E = $80 per room and 10 000 rooms rented each night, to A, at a market price of P_C = $100 per room and only 5000 rooms rented each night. A is, of course, on both the demand curve D and the new supply curve S₂. In this case, $100 is the demand price (P_C) of 5000 rooms—but in effect hotel owners receive only $60 (P_P), when you account for the fact that they have to collect and remit the $40 tax. From the point of view of hotel owners, it is as if they were on their original supply curve at point B.

Let’s check this again. How do we know that 5000 rooms will be supplied at a price of $100? Because the price net of tax (P_P = P_C − T) is $60, and according to the original supply curve, 5000 rooms will be supplied at a price of $60, as shown by point B in Figure 7-2.

Does this look familiar? It should. In Chapter 5 we described the effects of a quota on sales: a quota drives a wedge between the price paid by consumers and the price received by producers. An excise tax does the same thing. As a result of this wedge, consumers pay more (P_C) and producers receive less (P_P), with the gap being the tax wedge (T), so that P_C = P_P + T.

In our example, consumers—people who rent hotel rooms—end up paying $100 a night, $20 more than the pre-tax price of $80. At the same time, producers—the hotel owners—receive a price net of tax of $60 per room, $20 less than the pre-tax price. In addition, the tax creates missed opportunities: 5000 potential consumers who would have rented hotel rooms—those willing to pay $80 but not $100 per night—are discouraged from doing so. Correspondingly, 5000 rooms that would have been made available by hotel owners when they receive $80 are not offered when they receive only $60. Like a quota, this tax leads to inefficiency by distorting incentives and creating missed opportunities for mutually beneficial transactions.

It’s important to recognize that as we’ve described it, Potterville’s hotel tax is a tax on the hotel owners, not their guests—it’s a tax on the producers, not the consumers. Yet the price received by producers, net of tax (P_P), is down by only $20, half the amount of the tax, and the price paid by consumers (P_C) is up by $20. In effect, half the tax is actually being paid by consumers—the burden of the tax is shared between the suppliers and demanders.

What would happen if the city levied a tax on consumers instead of producers? That is, suppose that instead of requiring hotel owners to pay $40 a night for each room they rent, the city required hotel guests to pay $40 for each night they stayed in a hotel. The answer is shown in Figure 7-3. If a hotel guest must pay a tax of $40 per night, then the pre-tax price for a room paid by that guest must be reduced by $40 in order for the quantity of hotel rooms demanded post-tax to be the same as that demanded pre-tax. So the demand curve shifts downward by the amount of the tax from D₁ to D₂ = D₁ − T.

At every quantity demanded, the demand price—the price that consumers must be offered to demand a given quantity—has fallen by $40. This shifts the equilibrium from E to B, where the market price of hotel rooms is $60 and 5000 hotel rooms are bought and sold. In effect, hotel guests pay $100 when you include the tax. So from the point of view of guests, it is as if they were on their original demand curve at point A.

If you compare Figures 7-2 and 7-3, you will immediately notice that they show the same price effect. In each case, consumers pay an effective price of $100 (P_C), producers receive an effective price of $60 (P_P), and 5000 hotel rooms are bought and sold. In fact, it doesn’t matter who the government officially asks to pay the tax—whether it’s levied on suppliers or consumers—the equilibrium outcome is the same.

This insight illustrates a general principle of the economics of taxation: the incidence of a tax—who really bears the burden of the tax—is typically not a question you can answer by asking who writes the cheque to the government. In this particular case, a $40 tax on hotel rooms is reflected in a $20 increase in the price paid by consumers and a $20 decrease in the price received by producers. Here, regardless of whether the tax is levied on consumers or producers, the incidence of the tax is evenly split between them.¹

We’ve just learned that the economic incidence of an excise tax doesn’t depend on whom the government rules require to officially pay it. In the example shown in Figures 7-1 through 7-3, a tax on hotel rooms falls equally on consumers and producers, no matter who the tax is levied on. But it’s important to note that this 50–50 split between consumers and producers is a result of our assumptions in this example. In the real world, the incidence of an excise tax usually falls unevenly between consumers and producers, as one group bears more of the burden than the other.

What determines how the burden of an excise tax is allocated between consumers and producers? The answer depends on the shapes of the supply and the demand curves. More specifically, the incidence of an excise tax depends on the price elasticity of supply and the price elasticity of demand. We can see this by looking first at a case in which consumers pay most of an excise tax, then at a case in which producers pay most of the tax.

When an Excise Tax Is Paid Mainly by Consumers Figure 7-4 shows an excise tax that falls mainly on consumers: an excise tax on gasoline, which we set at $0.25 per litre. (There really is a federal excise tax on gasoline, though it is actually only $0.10 per litre in Canada. In addition, provinces, territories, and even several cities impose excise taxes between $0.09 and $0.24 per litre. Finally, the GST/HST and, if applicable, provincial sales taxes also apply to gasoline sales.) According to Figure 7-4, in the absence of the tax, gasoline would sell for $1.05 per litre.

Two key assumptions are reflected in the shapes of the supply and demand curves in Figure 7-4. First, the price elasticity of demand for gasoline is assumed to be very low, so the demand curve is relatively steep. Recall that a low price elasticity of demand means that the quantity demanded changes little in response to a change in price—a feature of a steep demand curve. Second, the price elasticity of supply of gasoline is assumed to be very high, so the supply curve is relatively flat. A high price elasticity of supply means that the quantity supplied changes a lot in response to a change in price—a feature of a relatively flat supply curve.

We have just learned that an excise tax drives a wedge, equal to the size of the tax, between the price paid by consumers and the price received by producers. This wedge drives the price paid by consumers up and the price received by producers down. But as we can see from Figure 7-4, in this case those two effects are very unequal in size. The price received by producers falls only slightly, from $1.05 to $1.00, but the price paid by consumers rises by a lot, from $1.05 to $1.25. In this case consumers bear the greater share of the tax burden.

This example illustrates another general principle of taxation: When the price elasticity of demand is low and the price elasticity of supply is high, the burden of an excise tax falls mainly on consumers. Why? A low price elasticity of demand means that consumers have few substitutes and so little alternative to buying higher-priced gasoline. In contrast, a high price elasticity of supply results from the fact that producers have many production substitutes for their gasoline (that is, other uses for the crude oil from which gasoline is refined). This gives producers much greater flexibility in refusing to accept lower prices for their gasoline. And, not surprisingly, the party with the least flexibility—in this case, consumers—gets stuck paying most of the tax. This is a good description of how the burden of the main excise taxes actually collected in Canada today, such as those on cigarettes and alcoholic beverages, is allocated between consumers and producers.

When an Excise Tax Is Paid Mainly by Producers Figure 7-5 shows an example of an excise tax paid mainly by producers, a $5.00 per day tax on downtown parking in a small city. In the absence of the tax, the market equilibrium price of parking is $6.00 per day.

We’ve assumed in this case that the price elasticity of supply is very low because the lots used for parking have very few alternative uses. This makes the supply curve for parking spaces relatively steep. The price elasticity of demand, however, is assumed to be high: consumers can easily switch from the downtown spaces to other parking spaces a few minutes’ walk from downtown, spaces that are not subject to the tax. This makes the demand curve relatively flat.

The tax drives a wedge between the price paid by consumers and the price received by producers. In this example, however, the tax causes the price paid by consumers to rise only slightly, from $6.00 to $6.50, but the price received by producers falls a lot, from $6.00 to $1.50. In the end, consumers bear only $0.50 of the $5.00 tax burden, with producers bearing the remaining $4.50.

Again, this example illustrates a general principle: When the price elasticity of demand is high and the price elasticity of supply is low, the burden of an excise tax falls mainly on producers. A real-world example is a tax on purchases of existing houses. Just before the recession hit Canada in 2008, the cash-strapped city of Toronto implemented a municipal land transfer tax to raise revenue. A land transfer tax is collected as a percentage of the price at which a residential or commercial property sells. To politicians, the idea was simple: homes sales were strong and in the preceding years prices had steadily risen. The tax rate was designed to rise somewhat as the price increased, so to some degree the higher the price paid the more tax paid in both relative and absolute terms. So besides helping the city meet its budgetary needs, the tax was intended to be paid mostly by well-off outsiders moving into desirable locations and purchasing homes from the less well-off original occupants. But this ignores the fact that the price elasticity of demand for houses in Toronto is often high, because potential buyers can choose to move to other towns. Furthermore, the price elasticity of supply is often low because most sellers must sell their houses due to job transfers or to provide funds for their retirement. So taxes on home purchases are actually paid mainly by the less well-off sellers—not, as Toronto officials may have imagined, by wealthy buyers.²

Putting It All Together We’ve just seen that when the price elasticity of supply is high and the price elasticity of demand is low, an excise tax falls mainly on consumers. And when the price elasticity of supply is low and the price elasticity of demand is high, an excise tax falls mainly on producers. This leads us to the general rule: When the price elasticity of demand is higher than the price elasticity of supply, an excise tax falls mainly on producers. When the price elasticity of supply is higher than the price elasticity of demand, an excise tax falls mainly on consumers. So it is elasticity—not the tax legislation—that determines the incidence of an excise tax.