Case Study 2 — The Welfare Cost of Cigarette Taxes

Cigarette taxes are one of the most heavily studied applications of the surplus framework. They're also one of the cleanest examples of how a tax with a substantial cost to consumers (in dollars per pack) can have a relatively small welfare cost (deadweight loss) — and how the framework's blind spot on equity makes the policy more contested than the efficiency calculation alone would suggest.

This case study walks through cigarette taxes using the consumer/producer surplus framework, then notes what the framework cannot capture and what that means for the policy debate.

The basic facts

The U.S. federal cigarette tax is $1.01/pack. State cigarette taxes range from $0.17 (in Missouri) to $5.35 (in New York City when the city's $1.50 is added to New York State's $4.35). The combined federal + state + local tax in some U.S. cities can exceed $7/pack. Internationally, the U.K., Norway, and Australia have even higher cigarette taxes.

The price of a pack of cigarettes in the United States ranges from about $7 (in low-tax states) to about $14+ (in NYC). Without taxes, the manufacturer's price would be roughly $4–$5/pack, depending on the brand.

The economic justification for cigarette taxes has two parts: (1) raise revenue, and (2) reduce smoking by raising the cost. Both are real, but they pull in opposite directions — if the tax dramatically reduces smoking, it raises less revenue per smoker than if smoking is unchanged. The right tax level depends on the elasticity of demand and the goal of the policy.

The standard surplus analysis

The empirical literature on cigarette demand consistently finds that the price elasticity of demand for cigarettes is small — typically in the range of −0.3 to −0.6 in the short run and somewhat larger in the long run. Demand is inelastic because of nicotine addiction.

Using these numbers, we can do a rough surplus calculation.

Without the tax: suppose the equilibrium price is $5/pack and equilibrium quantity is 100 packs/day in some hypothetical market. Consumer surplus and producer surplus together add up to some baseline total.

**With a $1.50 tax**: the supply curve shifts up by $1.50 (sellers now need to receive $1.50 more to produce the same quantity). The buyer's price rises (but by less than $1.50, because some of the tax is absorbed by sellers). The seller's price falls (but by less than $1.50). Quantity falls.

Because demand is inelastic (elasticity ≈ −0.4 in our example), most of the tax falls on consumers. The buyer's price might rise from $5 to $6.30 (an increase of $1.30 of the $1.50 tax), and the seller's price might fall from $5 to $4.80 (a decrease of $0.20). Total burden distribution: about 87% on consumers, 13% on producers.

Quantity falls modestly. With elasticity of −0.4 and a 26% buyer-price increase, the quantity should fall by about 10–11%. So Q goes from 100 packs/day to about 89 packs/day in this hypothetical.

Surplus changes: - Consumer surplus loss: significant (consumers pay $1.30 more per pack on the 89 they still buy + lose all the value on the 11 they no longer buy) - Producer surplus loss: smaller (sellers receive $0.20 less per pack) - Government revenue: $1.50 × 89 = $133.50/day (plus revenue on the inelastic-but-diminished quantity) - Deadweight loss: small (a small triangle, because Q didn't fall much)

The deadweight loss is small relative to the total surplus loss because the quantity reduction is small. This is the inelastic-good story: most of the welfare cost is transfer (from consumers to government), not destruction (deadweight loss).

This is one of the cleanest applications of the surplus framework, and it produces a clear policy implication: cigarette taxes are an efficient way to raise revenue (compared to taxes on more elastic goods, which would have more deadweight loss). They're also an effective way to reduce smoking (because of the inelastic but nonzero demand response).

What the framework cannot capture

The surplus calculation makes cigarette taxes look pretty good: small deadweight loss, substantial revenue, modest reduction in smoking. But the calculation has some serious blind spots.

Blind spot 1 — The composition of demand

Cigarette taxes fall heavily on the people who smoke. But who smokes? In modern America, smoking is concentrated among lower-income, less-educated, and disproportionately white-rural and black-urban populations. The cigarette tax is not a tax on "consumers" in the abstract — it's a tax on specific groups, mostly low-income.

This means cigarette taxes are regressive in a strong sense: a fixed-dollar tax represents a much larger percentage of income for poor smokers than for rich ones. The same $1.50 tax that's $1 of a $50,000 income is $4 of a $20,000 income (in proportional terms — it's a much bigger bite for the poorer person).

The surplus framework treats all consumer surplus the same: a dollar of welfare loss is a dollar of welfare loss, regardless of whose dollar it is. This is the "utilitarian" assumption. Most people, when pressed, would say a dollar means more to a poor person than to a rich one. The framework doesn't capture that intuition, but it matters morally.

Blind spot 2 — The smoking-as-addiction question

The surplus framework assumes that consumers are rational actors whose willingness to pay reflects their genuine preferences. For smokers, this is empirically dubious. Many smokers say they wish they didn't smoke. Many smokers say they would quit if they could. Behavioral economics has documented that smokers tend to underweight long-term health costs and overweight short-term enjoyment — the kind of present-biased preference reversal we previewed in Chapter 1 and will see in much more depth in Chapter 10.

If smokers' willingness to pay reflects an addiction-driven distortion of their preferences, then the surplus calculation may understate the welfare benefit of taxes. The smokers who are pushed to quit by the tax may, on net, be better off than the framework suggests — because their "willingness to pay" was distorted by addiction in the first place.

Some behavioral economists have argued that this means cigarette taxes can be welfare-improving even at levels that look suboptimal in the standard framework. The argument is contested but serious, and it's one of the places where behavioral economics is changing how policy economists think about taxes on addictive goods.

Blind spot 3 — Health externalities

Smoking imposes costs on people other than the smoker — secondhand smoke, healthcare costs (especially for those covered by public insurance), and the loss of productive years for the smoker themselves. These externalities are not captured in the standard surplus calculation, but they're often the primary economic justification for taxing tobacco. Pigouvian taxation (the externality framework we'll see in Chapter 11) provides the formal language: when smoking causes external costs, taxing it brings the private cost of smoking closer to the social cost, improving welfare.

The CDC estimates that smoking costs the U.S. about $300 billion per year in direct medical costs and lost productivity. If even part of this is genuinely external (not borne by the smoker), the case for taxes is strong on Pigouvian grounds.

Blind spot 4 — Black markets

When taxes get high enough, some smokers turn to black-market cigarettes — bootlegging from low-tax states, illegal imports, or counterfeit brands. New York City has had a notable black-market problem because of its very high taxes. The black market is a real form of avoidance that: - Reduces tax revenue - Doesn't reduce smoking as much as the official quantity decline suggests - Creates illegal activity and the costs of enforcing it - Can produce more dangerous (counterfeit) cigarettes

The surplus framework doesn't easily handle black markets. They show up as smaller-than-expected revenue and smaller-than-expected health benefits. They're a real reason to be cautious about very high cigarette taxes in some jurisdictions.

Blind spot 5 — The dignity question

Some people object to cigarette taxes on dignity grounds: they say smoking is a personal choice and the government should not be in the business of nudging people away from it through punitive pricing. The surplus framework cannot evaluate this objection — it doesn't take "dignity" or "freedom" as inputs. If you weight individual freedom heavily, the cigarette tax is paternalistic in a way the framework can't capture.

What economic analysis can and cannot tell you about cigarette taxes

The surplus framework tells you: - Cigarette taxes are an efficient way to raise revenue (small deadweight loss) - They reduce smoking (the empirical elasticity is small but nonzero) - The burden falls mostly on consumers (because demand is inelastic) - The total welfare cost is dominated by transfer to government, not destruction

The framework does not tell you: - Whether the regressive distribution is acceptable - Whether smokers' "willingness to pay" reflects genuine preferences or addiction - How to weight health externalities against private utility loss - Whether black markets cancel out the benefits at high tax levels - Whether paternalism is compatible with individual freedom

These are normative questions, and the framework cannot resolve them. Economists who support high cigarette taxes typically weight health externalities heavily, accept the regressivity as a cost worth paying, and discount the dignity objection. Economists who oppose them typically weight personal freedom and the regressive distribution more heavily. Both positions are defensible — and both are using the same surplus framework as their starting point.

Discussion questions

  1. The case study notes that cigarette taxes are regressive. Should this disqualify them as a policy, or can the regressivity be addressed by using the revenue in progressive ways?

  2. The behavioral economics argument is that smokers' willingness to pay is distorted by addiction. Do you find this argument persuasive? If so, what does it imply for how the surplus framework should be applied to addictive goods?

  3. New York City has very high cigarette taxes and a notable black market. At what point does a tax become "too high"? What's the framework for thinking about this?

  4. The surplus framework treats a dollar of consumer surplus as equivalent to a dollar of producer surplus or a dollar of government revenue. Is this assumption acceptable? Are there situations where it's clearly wrong?

  5. Suppose your state is considering raising the cigarette tax by $1/pack. Use the surplus framework to evaluate the proposal. What additional information would you want before recommending for or against?