Lecture 08

.title[
# Lecture 08
]
.subtitle[
## Carbon Tariffs
]
.author[
### Ivan Rudik
]
.date[
### AEM 4510
]

---

``` r
if (!require("pacman")) install.packages("pacman")
```

```
## Loading required package: pacman
```

``` r
pacman::p_load(
  tidyverse, xaringanExtra, rlang, patchwork
)
options(htmltools.dir.version = FALSE)
knitr::opts_hooks$set(fig.callout = function(options) {
  if (options$fig.callout) {
    options$echo <- FALSE
  }
knitr::opts_chunk$set(echo = TRUE, fig.align="center")
  options
})
```

```
## Warning in xaringanExtra::style_panelset(panel_tab_color_active = "red"): 'xaringanExtra::style_panelset' is deprecated.
## Use 'style_panelset_tabs' instead.
## See help("Deprecated")
```

```
## Warning in style_panelset_tabs(...): The argument names of `style_panelset()`
## changed in xaringanExtra 0.1.0. Please refer to the documentation to update to
## the latest names.
```

```
## NULL
```

---

# Roadmap

1. What are carbon tariffs?
2. Basic trade theory: supply, demand, and tariffs
3. Adding pollution externalities to trade
4. Unilateral carbon taxes and leakage
5. Carbon tariffs to prevent leakage
6. Carbon tariffs alone
7. World carbon taxes
8. Different pollution intensities across countries

---

# What are carbon tariffs?

---

# Carbon tariffs

Carbon tariffs are tariffs imposed on imports based on their .hi[carbon content]

Why would we want carbon tariffs?

1. .hi[Leakage:] if country `$i$` imposes a carbon tax, production may shift to country `$j$` with no carbon tax

2. .hi[Competitiveness:] domestic firms face higher costs than foreign competitors

Carbon tariffs aim to "level the playing field" and reduce leakage

---

# Carbon tariffs in practice

.hi[EU Carbon Border Adjustment Mechanism (CBAM):]
- Phasing in 2023-2026, full implementation 2026
- Covers cement, iron/steel, aluminum, fertilizers, electricity, hydrogen
- Importers must purchase certificates matching EU ETS price

---

# This lecture

We will use a simple partial equilibrium framework from Brunel and Levinson (2024) to understand:

1. How carbon tariffs work in theory

2. When they improve welfare

3. What complications arise in practice

The key insight: carbon tariffs are equivalent to a .hi[consumption tax] plus a .hi[production subsidy]

---

# Basic trade theory

---

# A simple trade model

Consider a small open economy (country `$i$`) that:
- Takes world prices as given
- Can import or export a homogeneous good

Key curves:
- `$S$`: domestic supply
- `$D$`: domestic demand
- `$p_w$`: world price

---

# Free trade equilibrium

```
## [Figure 1: Standard tariff - S, D curves, deadweight triangles a, b]
```

With free trade at price `$p_w$`:
- Domestic production: `$Q_1$`
- Domestic consumption: `$Q_4$`
- Imports: `$Q_4 - Q_1$`

---

# Adding a tariff

A tariff `$t$` raises the domestic price to `$p_w + t$`

Effects:
- Domestic production rises: `$Q_1 \rightarrow Q_2$`
- Domestic consumption falls: `$Q_4 \rightarrow Q_3$`
- Imports fall: `$(Q_4 - Q_1) \rightarrow (Q_3 - Q_2)$`

.hi[Deadweight loss:] triangles `$a + b$`
- `$a$`: production distortion (inefficient domestic production)
- `$b$`: consumption distortion (foregone consumer surplus)

---

# Tariff revenue

The tariff generates revenue: `$t \times (Q_3 - Q_2)$`

This is a transfer from consumers to the government

Net welfare effect of tariff: `$-(a + b)$` (deadweight loss)

Standard result: .hi[tariffs reduce welfare] in a small open economy

---

# Adding pollution

---

# Production creates pollution

Now suppose production of this good creates pollution

Let `$e$` = emissions per unit of output (same everywhere)

Let `$d$` = marginal damage from emissions (global pollutant)

The .hi[marginal social cost (MSC)] of production is: `$MSC = S + e \times d$`

MSC lies above the private supply curve by the external cost `$e \times d$`

---

# Free trade with pollution

```
## [Figure 2a: Country i with MSC curve above S]
```

Country `$i$`'s MSC curve lies above its supply curve

The gap is the external cost: `$e \times d$`

---

# Global perspective

```
## [Figure 2b: World S, MSC, D curves]
```

From a global perspective:
- Free trade output: where `$D = S$` (too much!)
- Efficient output: where `$D = MSC$` (less production)

---

# The externality problem

With a global pollutant like `$CO_2$`:

- Damages occur regardless of where emissions happen
- No country has incentive to fully internalize the externality
- Free trade leads to .hi[too much production] globally

A global carbon tax `$= e \times d$` would achieve efficiency

But what if only .hi[one country] acts?

---

# Unilateral carbon tax

---

# Country `$i$` imposes a carbon tax

Suppose country `$i$` imposes a carbon tax `$\tau = e \times d$` on domestic production

This shifts country `$i$`'s supply curve up by `$\tau$`

But the .hi[world price doesn't change] (small open economy)

---

# Effects of unilateral carbon tax

```
## [Figure 3: Unilateral carbon tax with leakage]
```

---

# Unilateral carbon tax effects

Country `$i$`'s carbon tax:
- Reduces domestic production
- Does not change domestic consumption (price unchanged)
- .hi[Increases imports]

The increase in imports means .hi[foreign production increases]

This is .hi[carbon leakage]: emissions shift abroad rather than being reduced

---

# Leakage undermines the tax

With identical carbon intensities:
- Every unit of reduced domestic production = 1 unit more foreign production
- .hi[Net global emissions unchanged!]

The carbon tax:
- Generates tax revenue `$R_t$` (rectangle)
- Creates deadweight loss `$d$` (triangle) from production distortion
- Achieves .hi[zero environmental benefit]

Country `$i$` bears costs but gets no environmental gain

---

# Carbon tax + carbon tariff

---

# Adding a carbon tariff

Now suppose country `$i$` adds a carbon tariff equal to the carbon tax: `$t = \tau = e \times d$`

This tariff applies to the carbon content of imports

Domestic price rises to: `$p_w + t$`

---

# Carbon tax + tariff effects

```
## [Figure 4: Carbon tax + tariff effects]
```

---

# How the tariff helps

The carbon tariff:
- Raises the domestic price
- .hi[Reduces consumption] in country `$i$`
- Reduces imports (less leakage)

Net effect:
- Production distortion from tax alone: `$-d$`
- Consumption distortion from tariff: `$-b$`
- .hi[Environmental benefit:] reduced global emissions

If environmental benefit `$> d + b$`: net welfare gain!

---

# The key insight

With identical carbon intensities everywhere:

Why?
- Tax on domestic production raises domestic costs
- Tariff on imports raises import costs
- Both domestic and imported goods face the same carbon price
- Equivalent to taxing consumption of the carbon-intensive good

---

# Carbon tariff alone

---

# What if we only use a tariff?

Some proposals call for carbon tariffs .hi[without] a domestic carbon tax

What does this accomplish?

---

# Carbon tariff alone

```
## [Figure 5: Carbon tariff alone = consumption tax + production subsidy]
```

---

# Tariff alone = consumption tax + production subsidy

A carbon tariff without a domestic carbon tax is equivalent to:

1. .hi[Consumption tax:] raises price consumers pay

2. .hi[Production subsidy:] domestic producers don't pay the carbon cost

This .hi[increases] domestic production while reducing consumption

Domestic emissions may .hi[rise]!

---

# Is a tariff alone good policy?

A carbon tariff alone:
- Reduces imports (good for leakage)
- But .hi[subsidizes] domestic carbon-intensive production (bad)

Net environmental effect is ambiguous

Generally .hi[not as effective] as carbon tax + tariff

May violate WTO rules (discriminates against foreign producers)

---

# World carbon tax

---

# What if everyone taxes carbon?

Suppose all countries impose the same carbon tax `$\tau = e \times d$`

Now there's no competitive disadvantage

No leakage: all production faces the same carbon price

---

# World carbon tax

```
## [Figure 6: World carbon tax - efficient outcome]
```

---

# Efficient outcome

With a world carbon tax at `$\tau = e \times d$`:
- Global production equals efficient level
- Price reflects full social cost
- .hi[No need for carbon tariffs]

Carbon tariffs are a .hi[second-best] policy for when global coordination fails

---

# Insufficient world carbon tax

What if the world carbon tax is .hi[too low]?

Suppose world tax = `$\tau' < e \times d$`

Global emissions still too high, but less so than with no tax

---

# Too-small world tax

```
## [Figure 7: Too-small world carbon tax]
```

Can country `$i$` improve things with a tariff?

---

# Tariff on top of insufficient world tax

```
## [Figure 8: Overshooting with tariff on insufficient world tax]
```

---

# Overshooting problem

If country `$i$` adds a tariff `$= e \times d - \tau'$` on top of world tax `$\tau'$`:

- Domestic price rises to efficient level
- But .hi[domestic consumption] falls below efficient level
- Creates new distortion

Key issue: country `$i$` can only affect its .hi[own] consumption, not world production

A tariff can't fix insufficient global carbon pricing

---

# Different carbon intensities

---

# What if countries differ?

So far: same carbon intensity `$e$` everywhere

Reality: countries have very different carbon intensities
- Cleaner electricity grids
- Different production technologies
- Different regulatory environments

How should tariffs account for this?

---

# Cleaner domestic production

Suppose country `$i$` has lower carbon intensity than the rest of the world:
- `$e_i < e_j$` (domestic production is cleaner)

Now the .hi[efficient] policy differs for domestic vs. foreign goods

---

# Tariff based on domestic carbon intensity

```
## [Figure 9: Tariff based on domestic carbon intensity]
```

If tariff = domestic carbon tax = `$e_i \times d$`:
- Correctly prices domestic emissions
- .hi[Under-prices] foreign emissions (since `$e_j > e_i$`)

---

# Tariff based on foreign carbon intensity

```
## [Figure 10: Tariff based on foreign carbon intensity]
```

If tariff = `$e_j \times d$` (foreign intensity):
- Correctly prices foreign emissions
- But tariff `$>$` domestic tax: .hi[looks protectionist]

---

# The measurement problem

Setting tariffs based on actual foreign carbon intensity requires:

1. Measuring emissions in foreign production
2. Verifying foreign carbon prices
3. Adjusting for different products/producers

This is .hi[extremely difficult] in practice

EU CBAM uses default values, allows importers to prove lower emissions

---

# Complications

---

# Real-world complications

Several factors complicate carbon tariff design:

1. .hi[Resource shuffling:] clean production exported, dirty production consumed domestically

2. .hi[Non-price policies:] regulations vs. carbon prices

3. .hi[Export rebates:] should exports get carbon tax rebates?

4. .hi[Retaliation:] trading partners may impose counter-tariffs

---

# Resource shuffling

Suppose country `$j$` has both clean and dirty production

With a carbon tariff based on average intensity:
- Export the clean stuff to country `$i$`
- Consume the dirty stuff domestically

No change in country `$j$`'s total emissions!

This is .hi[another form of leakage]

---

# Non-price policies

Many countries use regulations rather than carbon prices:
- Fuel economy standards
- Renewable portfolio standards
- Building codes

How do you calculate the "equivalent carbon price"?

Very difficult: may require complex modeling

---

# Export rebates

Should country `$i$` rebate carbon taxes on exports?

Arguments for:
- Maintains competitiveness
- Symmetric with import tariffs

Arguments against:
- Reduces domestic incentive to decarbonize
- May look like an export subsidy (WTO issues)

---

# WTO compatibility

Carbon tariffs face WTO scrutiny:

1. .hi[National treatment:] can't discriminate against foreign goods

2. .hi[Most favored nation:] can't discriminate among trading partners

3. .hi[GATT Article XX:] exceptions for environmental protection

EU CBAM designed carefully to fit within WTO rules

---

# Key takeaways

---

# Summary

1. .hi[Carbon tariffs] aim to address leakage and competitiveness concerns

2. .hi[Carbon tax + tariff] is equivalent to a consumption tax
   - Reduces consumption of carbon-intensive goods
   - Can improve welfare if environmental benefit exceeds distortion costs

3. .hi[Carbon tariff alone] = consumption tax + production subsidy
   - May increase domestic emissions
   - Less effective than tax + tariff

---

# Summary (continued)

4. .hi[World carbon tax] is first-best
   - Carbon tariffs are second-best when global coordination fails

5. .hi[Different carbon intensities] create measurement challenges
   - Setting tariff = foreign intensity is efficient but looks protectionist
   - Setting tariff = domestic intensity under-prices foreign emissions

6. .hi[Many complications] in practice
   - Resource shuffling, non-price policies, export rebates, WTO rules

---

# The bottom line

Carbon tariffs are a .hi[complement] to carbon pricing, not a substitute

They work best when:
- Combined with a domestic carbon tax
- Set based on actual carbon content
- Part of broader international coordination

Alone, they are an imperfect tool that may create new distortions

---

# References

Brunel, C. and A. Levinson (2024). "Carbon Tariffs 101." NBER Working Paper 33024.

EU Carbon Border Adjustment Mechanism: https://taxation-customs.ec.europa.eu/carbon-border-adjustment-mechanism_en