Lecture 9: Taxation and Regulation

Learning Objectives

By the end of today you should be able to:

Explain what deadweight loss is and why taxes create it
Distinguish statutory from economic incidence of a tax
Use elasticity intuition to predict who bears a tax burden
Analyze price floors and price ceilings with supply and demand
Evaluate the tradeoffs in labor market regulations like the minimum wage

Where We Are

Last lecture: market power → firms restrict output → deadweight loss

Today: government taxes and regulations can do the same thing — intentionally

The policy question is never “does intervention have costs?” It’s “are the benefits worth the costs?”

Taxes Reduce Surplus — But How Much?

Taxes:

Raise prices for buyers → reduce consumer surplus
Lower prices for sellers → reduce producer surplus
Generate revenue for government

Government revenue is a transfer — society doesn’t lose it; it’s redistributed

But the trades that don’t happen because of the tax — those are a real loss

This is the Deadweight Loss (DWL), also called excess burden

Watching the Tax Wedge Form

Figure 1: A per-unit tax shifts the effective demand curve down: CS and PS shrink as tax revenue and DWL emerge

Before and After: The $4 Tax

Figure 2: Left: competitive equilibrium — full CS and PS, no DWL. Right: after a $4 per-unit tax — CS and PS shrink, revenue and DWL emerge.

A Tax Creates a Price Wedge

Figure 3: A $4 per-unit tax drives a wedge between what buyers pay and what sellers receive

What Changed?

Before tax:

$Q^* = 6$, $P^* = \$8$
CS + PS = total surplus
No DWL

After $4 tax:

$Q_{tax} = 4$
$P_{buyer} = \$10$ ↑ by $2
$P_{seller} = \$6$ ↓ by $2
Tax revenue: $\$4 \times 4 = \$16$

Each side bears exactly half because slopes are equal.

But that is not always the case — it depends on elasticities.

A Case Study: The British Window Tax

Britain taxed windows in homes from 1696 to 1851

Windows	1747–57 rate	1761–65 rate
Fewer than 8	Free	Free
8–9	Free	1 shilling
10–14	6 pence/window	1s 6p/window
15–19	9 pence/window	1s 9p/window
20+	1 shilling/window	2 shillings/window

Why windows? Hard to hide. Meant as a wealth proxy — like an ability-to-pay tax.

What Happened?

People responded to the incentive:

Bricked up windows in existing homes
New houses built with fewer windows
Working-class tenements: no light

Health consequences:

Poor air circulation → disease spread
Dysentery, typhus, gangrene
1848 report: “houses of darkness”
Led to repeal of the tax in 1851

Bricked-up windows still visible on historic British buildings

“Daylight robbery” entered the language here

Window Thresholds

Histogram: Number of Windows by Year Home Built
(Oates & Schwab 2015, Journal of Economic Perspectives)

Quantifying the DWL

Oates & Schwab (2015, Journal of Economic Perspectives) estimated:

DWL ≈ 13.4% of tax revenue (a lower bound)
Marginal excess burden ≈ $0.23 per additional dollar raised

Meaning: for every $1 the government collected, society lost an additional $0.23 in destroyed value — windows that were bricked up, houses built darker, health costs from poor ventilation.

“The tax on windows renders light dearer.” — Adam Smith observed the perverse incentive decades before repeal.

The Harberger Triangle

Figure 4: Starting from competitive equilibrium, a rising tax shifts the effective demand down, shrinking CS and PS while filling in tax revenue and DWL

Why DWL Grows So Fast

\[\text{DWL} = \frac{1}{2} \times \underbrace{\Delta Q}_{\text{quantity lost}} \times \underbrace{\Delta P}_{\text{price wedge (the tax)}}\]

This is the area of the Harberger triangle.

Both $\Delta Q$ and $\Delta P$ grow with the tax rate $t$

\[\Rightarrow \text{DWL} \propto t^2\]

Tip

Policy implication: A 20% tax has 4× the DWL of a 10% tax — not 2×.

It is more efficient to tax many goods at low rates than one good at a high rate.

DWL Grows as Demand Becomes Elastic

Figure 5: A $4 tax on a market with supply P=2+Q. As demand becomes more elastic (flatter), the DWL triangle expands rapidly while tax revenue shrinks.

DWL by Elasticity: Side-by-Side Comparison

Figure 6: Same $4 tax, same supply curve — four demand elasticities on identical axes. DWL grows rapidly as demand becomes elastic; revenue falls.

Who Pays for Tariffs?

The Trump administration raised tariffs on imports (2018, renewed 2025).

Administration’s claim: foreign producers pay
Critics’ claim: U.S. consumers pay via higher prices

Both are using “statutory incidence” thinking.

Economics gives a clearer answer — but it requires elasticities

Statutory vs. Economic Incidence

Statutory incidence: who is legally required to send the check to the government

Economic incidence: how prices change relative to the no-tax world — who actually bears the burden

Important

The key result: Statutory incidence is irrelevant in standard models.

Whether the buyer or seller writes the check has no effect on who actually bears the burden.

What matters: elasticities

Statutory Incidence Is Irrelevant: The Proof

Figure 7: Whether the $4 tax is levied on buyers or sellers, the economic outcome is identical: same P_buyer, P_seller, quantity, revenue, and DWL

Elasticity Determines Who Pays

Figure 8: Same $2 tax, same equilibrium (Q=4, P=6), very different incidence. Incidence percentages: share of tax/tax.

Elasticity Shifts the Burden: Demand vs. Supply

Figure 9: Left: varying demand elasticity (supply fixed). Right: varying supply elasticity (demand fixed). In both cases, the less elastic side bears more of the burden.

Interactive: Explore Tax Incidence Yourself

You can interactively explore tax incidence below:

Adjust demand slope, supply slope, and tax size with the sliders to see how incidence and deadweight loss change.

The Intuition

Inelastic demand (steep curve):

Buyers really want the good
They can’t easily substitute away
Sellers pass the tax on to buyers
Buyers bear most of the burden

Elastic demand (flat curve):

Buyers have good substitutes
They walk away at higher prices
Sellers must absorb the tax
Sellers bear most of the burden

Rule: The less elastic side of the market bears more of the tax burden.

Special Cases

Scenario	Who bears the tax?	Example
Perfectly inelastic demand	Buyers (100%)	Insulin
Perfectly elastic demand	Sellers (100%)	Yellow M&M’s
Perfectly inelastic supply	Sellers (100%)	Helium
Perfectly elastic supply	Buyers (100%)	Price-taker

Of course, these are extremes and exceptions exist, but for most goods imported from global markets, foreign supply is highly elastic — sellers can sell elsewhere.

→ Tariff incidence falls on domestic buyers.

Back to Tariffs

Amiti, Redding & Weinstein (2019) studied the 2018 Trump tariffs:

U.S. import prices rose nearly one-for-one with the tariff rate
Foreign export prices barely changed
Conclusion: U.S. consumers and importers bore essentially the full burden

The administration’s claim was economically wrong — not because tariffs are bad policy per se, but because statutory incidence ≠ economic incidence.

Governments Can Also Set Prices Directly

Instead of a tax, government can mandate a minimum or maximum price.

Policy	Binds when…	Creates…
Price floor	set above equilibrium	surplus (excess supply)
Price ceiling	set below equilibrium	shortage (excess demand)

Both create a gap between quantity demanded and quantity supplied — and both create DWL for exactly the same reason as a tax.

The Minimum Wage: A Price Floor in the Labor Market

Figure 10: Minimum wage above equilibrium creates a surplus of labor (unemployment)

Who Wins and Who Loses?

Winners:

Workers who keep their jobs: ✅ higher wages
Workers whose WTA was below $12: ✅ newly enter the market

Losers:

Workers who lose jobs or can’t find them: ❌ earn $0 instead of $9.33
Employers: 📉 higher costs, fewer hires

How big is the unemployment effect?

Depends on the elasticity of labor demand:

Inelastic (e.g., healthcare workers): small employment drop
Elastic (e.g., easily automated tasks): large employment drop

A $12 minimum wage has different effects on hospital nurses vs. fast-food cashiers.

Does the Minimum Wage Kill Jobs? The Evidence

The competitive model predicts unemployment. But the real world is more complicated.

Card & Krueger (1994): Fast-food employment in NJ increased after a minimum wage hike, relative to Pennsylvania (which didn’t raise it).

Why? Labor markets are often not perfectly competitive.

Employers may have monopsony power — wage-setting power on the buyer’s side — which suppresses wages below the competitive level. A minimum wage can correct this, increasing both wages and employment.

Caution

Key lesson: The competitive model gives one prediction, but the correct prediction depends on market structure. Elasticity and market power both matter.

Regulations as Implicit Taxes

Many labor regulations function like a per-unit tax on employment.

Overtime pay requirements (40-hour work week)
Employer-sponsored health insurance mandates (ACA)
Mandatory paid parental leave

Each raises the effective cost of a worker beyond the posted wage.

In a competitive labor market, the incidence analysis applies:

Workers value the benefit → they accept lower wages to compensate
If wages adjust fully, employment is unchanged

But adjustment may be incomplete — especially when mandates benefit some workers more than others.

Mandatory Parental Leave: A Tricky Tradeoff

Parental leave is valuable — but who bears the cost?

In theory (competitive market, perfect information):

Workers value leave → accept lower wages
Cost is passed back to workers; no employment effect

In practice:

Leave is valued more by some demographic groups (e.g. gender)
Employers may anticipate this and “statistically” discriminate
Even with anti-discrimination law, this is hard to observe

Note

Sweden’s solution: “father quotas” — portion of leave is reserved for fathers and is lost if unused.

Putting It Together

Intervention	Price wedge?	DWL?	Incidence?
Per-unit tax	✅	✅	Less elastic side
Price floor (min. wage)	✅	✅	Depends on elasticity
Price ceiling (rent control)	✅	✅	Depends on elasticity
Lump-sum tax	❌	❌	Whoever policy targets
Regulation (benefit mandate)	✅ (implicit)	✅	Depends on who values benefit

Taxes and regulations are not inherently bad — they fund public goods, redistribute income, and address market failures. But they carry real efficiency costs that grow with elasticity and scale.

Summary

Statutory incidence is irrelevant — elasticities determine who really pays
The less elastic side bears more of the tax burden — intuition first, formula follows
Taxes create deadweight loss — mutually beneficial trades that don’t happen because of the price wedge
DWL grows with $t^2$ — better to spread taxes widely at low rates than one good at a high rate; the window tax is a vivid real-world example
Price floors (minimum wage) create labor surpluses in competitive models — but empirical effects depend heavily on elasticity and market power
Regulations function as implicit taxes — incidence follows the same logic, but distributional heterogeneity complicates the picture

Next Time

Externalities and Public Goods

When markets fail — not just get distorted
Pollution, climate change, public goods, information asymmetries
When the DWL from intervention can be less than the DWL from the market failure itself