Econ 551 Government Finance: Revenues Fall 2019 Given by Kevin - - PowerPoint PPT Presentation

ECON 551: Lecture 12 1 of 55

Econ 551 Government Finance: Revenues Fall 2019

Given by Kevin Milligan Vancouver School of Economics University of British Columbia Lecture 12: Environmental Taxation

ECON 551: Lecture 12 2 of 55

Agenda

• 1. Overview
• 2. Five methods of pollution abatement
• 3. Double Dividend Debate
• 4. Choice of Instruments
• 5. Canadian Policy Developments

ECON 551: Lecture 12 3 of 55

An ongoing Initiative: Ecofiscal Commission  5-year limited-term thinktank—just ending now!  Focused on pricing externalities.  Brings together group of economists and former politicians.

http://youtu.be/RBNPHv0R-aw

ECON 551: Lecture 12 4 of 55

2 4 6 8 10 12 14 16 18 TUR SVN ISR KOR NLD CHL IRL ZAF EST DNK CZE CHN HUN GBR FIN ITA PRT CHE POL SVK LUX ISL AUT DEU SWE NOR BEL ESP FRA NZL CAN USA Percentage

Environment taxes as % of total tax revenue 2012

Source: OECD http://stats.oecd.org

ECON 551: Lecture 12 5 of 55

Agenda

• 1. Overview
• 2. Five methods of pollution abatement
• 3. Double Dividend Debate
• 4. Choice of Instruments
• 5. Canadian Policy Developments

ECON 551: Lecture 12 6 of 55

Setup for studying externalities

Here’s the common framework we will use:  2 consumers, h=1,2  Wealth 𝑥ℎ (no income; endowment economy)  Goods indexed by j: 𝑦1

ℎ, … , 𝑦𝐾 ℎ

 Prices 𝑞1, … , 𝑞𝐾  Externality a, results from action of consumer 1 Indirect utility function: 𝑊ℎ(𝑞, 𝑥ℎ, 𝑏) = max

𝑦ℎ 𝑉ℎ(𝑦ℎ, 𝑏) 𝑡. 𝑢. 𝑞 ∙ 𝑦ℎ ≤ 𝑥ℎ

ECON 551: Lecture 12 7 of 55

Setup for studying externalities

To keep the model as simple as possible, let’s eliminate income effects with a quasi- linear indirect utility function; assume price-taker. 𝑊ℎ(𝑞, 𝑥ℎ, 𝑏) = 𝜚(𝑏) + 𝑥ℎ Assume externality is negative, utility is concave: 𝜖𝜚1 𝜖𝑏 > 0 , 𝜖𝜚2 𝜖𝑏 < 0 , 𝜖2𝜚ℎ 𝜖𝑏2 < 0 Consumer 1’s problem: max

𝑏

𝜚1(𝑏) + 𝑥1 ⇒ 𝜖𝜚1 𝜖𝑏 (𝑏∗) = 0 Consumer 2 just eats his 𝑦𝑘

2’s. Nothing he can do about a.

ECON 551: Lecture 12 8 of 55

Five potential solutions to externality problem

• a. Planner’s solution
• b. Effluent charge / Pigouvian tax
• c. Quota / regulation / Command and Control
• d. Property rights
• e. Tradeable permits.

ECON 551: Lecture 12 9 of 55

Planner’s solution (a):

First, assume a planner…  No prices.  Planner just assigns allocations based on maximizing social welfare.  Let’s just use utilitarian SWF; add up the individual utilities. max

𝑏

𝜚1(𝑏) + 𝜚2(𝑏) ⇒ 𝜖𝜚1 𝜖𝑏 + 𝜖𝜚2 𝜖𝑏 = 0 ⇒ 𝜖𝜚1 𝜖𝑏 (𝑏 ̂) = − 𝜖𝜚2 𝜖𝑏 (𝑏 ̂)

ECON 551: Lecture 12 10 of 55

Planner’s solution (a):

 𝑏 ̂ is the efficient outcome.  𝑏∗ is the competitive outcome.

𝑏 𝑏∗ 𝑏 ̂ − 𝜖𝜚2 𝜖𝑏 𝜖𝜚1 𝜖𝑏

ECON 551: Lecture 12 11 of 55

Effluent charge / Pigouvian tax (b):

 Arthur Cecil Pigou, 1877 – 1959  University of Cambridge economist  His book The Economics of Welfare (1920) argued for a tax to account for social cost of goods with externalities.

ECON 551: Lecture 12 12 of 55

Effluent charge / Pigouvian tax (b):

Mandate a tax 𝑢𝑏 per unit of a. Set tax at rate: 𝑢𝑏 = −

𝜖𝜚2 𝜖𝑏 (𝑏

̂). Consumer 1 now solves: max

𝑏

𝜚1(𝑏) − 𝑢𝑏𝑏 ⇒ 𝜖𝜚1 𝜖𝑏 − 𝑢𝑏 = 0 ⇒ 𝜖𝜚1 𝜖𝑏 = 𝑢𝑏 = − 𝜖𝜚2 𝜖𝑏 (𝑏 ̂) We end up right at 𝑏 ̂, the efficient outcome.

ECON 551: Lecture 12 13 of 55

Effluent charge / Pigouvian tax (b):

Comments:  For now, assume revenue is returned as lump sum transfer – goal of tax isn’t to raise

• revenue. (We’ll deal with ‘double dividend’ later…)

 Why does this work? Tax forces consumer 1 to internalize the externality.  Note the high information requirement. Have to know costs and benefits.

 Note the centralization of the solution.

𝑏 𝑏∗ 𝑏 ̂ − 𝜖𝜚2 𝜖𝑏 𝜖𝜚1 𝜖𝑏

t t

ECON 551: Lecture 12 14 of 55

Quota / regulation / Command and Control (c):

Another solution is to simply choose a quota. Set it at 𝑏 ̂. You’re done. This is essentially the same thing as the Pigou solution, since we assume we know the costs and the benefits.  But we’re choosing the quantity rather than choosing the prices.

𝑏 𝑏 ̂ − 𝜖𝜚2 𝜖𝑏 𝜖𝜚1 𝜖𝑏

ECON 551: Lecture 12 15 of 55

Property rights solution (d):

The next solution solves the problem essentially by assuming it away – say that there were property rights that could be assigned and that there was a market for the good. Let’s try giving Consumer 2 the property rights. This means that he can shut guy 1 down to zero if he wants.  He makes a take-it-or-leave-it offer to consumer 1 consisting of a transfer T.  Consumer 1 will accept the offer iff 𝜚1(𝑏) − 𝑈 ≥ 𝜚1(0) Consumer 2’s problem is therefore: max

𝑏,𝑈 𝜚2(𝑏) + 𝑈 𝑡. 𝑢. 𝜚1(𝑏) − 𝑈 = 𝜚1(0)

(This assumes constraint is binding.)

ECON 551: Lecture 12 16 of 55

Property rights solution (d):

Substitute out the constraint for T, leaving us with: max

𝑏,𝑈 𝜚2(𝑏) + 𝜚1(𝑏) − 𝜚1(0)

The FOC for this problem is: 𝜖𝜚2 𝜖𝑏 + 𝜖𝜚1 𝜖𝑏 − 0 = 0 (note that 𝜚1(0) is a constant, so its derivative is zero). This simply returns us to the social optimum: ⇒ 𝜖𝜚1 𝜖𝑏 (𝑏 ̂) = − 𝜖𝜚2 𝜖𝑏 (𝑏 ̂)

ECON 551: Lecture 12 17 of 55

Property rights solution (d):

 Say Consumer 2 has the right to stop a. Consumer 1 can offer to pay transfer T (green arrow)  If consumer 1 chooses some a, T has to be paid to consumer 2.  T is the transfer demanded by Consumer 1. Consumer 1 not happy!  After point 𝑏 ̂, the amount Consumer 1 is willing to pay is no longer enough to compensate Consumer 2. So, production stops at 𝑏 ̂, the socially efficient amount! 𝑏 𝑏 ̂ − 𝜖𝜚2 𝜖𝑏 𝜖𝜚1 𝜖𝑏

ECON 551: Lecture 12 18 of 55

Property rights solution (d):

Comments:  Note the low information requirement here. No omniscient planner who can see the value of the externality. Just assign property rights.  Of course, this just defines the problem away. If an externality is a missing market caused by no property rights, then there will be no externality if there are property rights.  Value of this approach is to emphasize that a) property rights are one important cause of missing markets and b) we don’t necessarily have to rely on a central government solution.

ECON 551: Lecture 12 19 of 55

Property rights solution (d): Coase extension

 Now let’s reverse things so that consumer 1 has property rights.  So, Consumer 2 has to make a transfer to consumer 1 or else Consumer 1 will just do what he pleases – which is to do a*.

𝑏 𝑏 ̂ − 𝜖𝜚2 𝜖𝑏 𝜖𝜚1 𝜖𝑏

ECON 551: Lecture 12 20 of 55

You can show using the same set up that we end up back at: 𝜖𝜚1 𝜖𝑏 (𝑏 ̂) = − 𝜖𝜚2 𝜖𝑏 (𝑏 ̂) No matter which guy had the property rights, we got to the same outcome. We’ll discuss this at length soon.

𝑏 𝑏 ̂ − 𝜖𝜚2 𝜖𝑏 𝜖𝜚1 𝜖𝑏

ECON 551: Lecture 12 21 of 55

Property rights solution (d): Coase extension

Ronald Coase (Nobel 1991) “The Problem of Social Cost” (1960 Journal of Law and Economics) 31,000+ Google Scholar citations… “With costless market transactions, the decision of the courts concerning liability for damage would be without effect on the allocation of resources.” (p. 10) “But the ultimate result (which maximizes the value of production) is independent of the legal position if the pricing system is assumed to work without cost.” (Coase 1960, p. 8) Emphasizes a central role for transaction costs and property rights for understanding the impact of externalities.

ECON 551: Lecture 12 22 of 55

Property rights solution (d): Coase extension

Now let’s put my own words on it and try to state the ‘Coase Theorem.’ Strong form: If property rights are well-defined and bargaining costs are zero, then agents will achieve, through voluntary transactions, allocations that are Pareto efficient and invariant to the initial allocation of property rights. Weak form: If property rights are well-defined and bargaining costs are zero, then agents will achieve, through voluntary transactions, allocations that are Pareto efficient.  The difference here is the ‘invariant’ part. Example: Trains generating sparks that started fires in farm fields.  Who has the property rights? What happens if rights reversed? Huge impact on understanding the economic implications of legal matters.

ECON 551: Lecture 12 23 of 55

Property rights solution (d): Coase extension

But, Coase can fall apart:  Uncertainty in bargaining. You don’t know everybody’s cost and benefit functions.  Compensation causes over-investment. In anticipation that you will get a transfer, you change your decision about investment.  Free-riding in negotiation. There may be gains to be had, but co-ordination among beneficiaries may be difficult. (Dixit and Olson 2000)  Endowment effects: people behave differently depending on initial endowments (Thaler…)

ECON 551: Lecture 12 24 of 55

(e) Tradable permit: Canadian invention!

University of Toronto Professor John Dales published a book on the idea in 1968.  It was radical at the time. (More here and here.)  Taken up by environmental economists in the 1970s; put in practice in 1980s. Lesson: Ideas are powerful. Don’t be afraid to think of radical “impractical” ideas!

ECON 551: Lecture 12 25 of 55

Tradable permit (e):

Again, this solution defines the problem away.  The Arrovian problem is that there is a missing market.  The solution, therefore, is to make that market reappear. One way to do so for some circumstances is to introduce a tradeable permit.  This gives someone the right to emit a certain amount of pollution.  By varying the number of permits that are out there, we can vary the amount of pollution. Here’s how it works: You must pay Pa in order to have the right to emit one unit of a. Consumer 2 doesn’t like a, but he can issue some permits to allow a.

ECON 551: Lecture 12 26 of 55

Tradable permit (e):

Consumer 1 now solves: max

𝑏

𝜚1(𝑏) − 𝑞𝑏 ⇒ 𝜖𝜚1 𝜖𝑏 (𝑏 ̅) = 𝑞𝑏 Consumer 2 now solves: max

𝑏

𝜚2(𝑏) + 𝑞𝑏 ⇒ 𝜖𝜚2 𝜖𝑏 (𝑏 ̅) = −𝑞𝑏 In a market, the price pa should set demand equal to supply: 𝜖𝜚1 𝜖𝑏 (𝑏 ̅) = 𝑞𝑏 = − 𝜖𝜚2 𝜖𝑏 (𝑏 ̅)

ECON 551: Lecture 12 27 of 55

(e) Tradable permit:

Consumer 1 has to pay for each permit. How much for how many?  Consumer 1 is willing to pay up to the marginal benefit for each permit.  Consumer 2 is willing to accept up to the marginal cost of pollution for each permit.  Equilibrium is at 𝑏 ̂.

𝑏 𝑏 ̂ − 𝜖𝜚2 𝜖𝑏

= supply of permits

𝜖𝜚1 𝜖𝑏

= demand for permits

Pa

ECON 551: Lecture 12 28 of 55

Tradable permit (e):

Comments:  Again, this is the efficient solution.  It is decentralized.  All you need to do is to be able to legally enforce the permits – monitor and penalize if someone exceeds.  Enforcement is harder in the international context. In other circumstances, monitoring is hard.

ECON 551: Lecture 12 29 of 55

Agenda

• 1. Overview
• 2. Five methods of pollution abatement
• 3. Double Dividend Debate
• 4. Choice of Instruments
• 5. Canadian Policy Developments

ECON 551: Lecture 12 30 of 55

The Double Dividend Debate

With a carbon tax or tradable permits, here’s the double dividend claim:  We have less pollution: that’s a dividend!  We get some extra revenue we can use to lower other taxes: that’s a dividend! But, remember back to Goulder and Williams (2003)? Equation (16)

1 𝜇 𝑒𝑉 𝑒𝜐𝐷𝑙

=

𝜐𝐷𝑙𝐷𝑙 𝑞𝐷𝑙

𝜁𝐷𝑙 + ∑

𝜐𝐷𝑗𝐷𝑗 𝑞𝐷𝑙 𝑗≠𝑙

𝜁𝐷𝑗𝐷𝑙 + ∑

𝜐𝐽𝑘𝐽𝑘 𝑞𝐷𝑙

𝜁𝐽𝑘𝐷𝑙 −

𝜐𝑀𝑀 𝑞𝐷𝑙 𝑂 𝑘=1

𝜁𝑀𝐷𝑙 First term:

• wn-price effect—the traditional Harberger triangle

Second term: cross-price impact on other consumption goods. Only matters if that good is taxed a lot so that 𝜐𝐷𝑗𝐷𝑗 is big, or strong substitute/complement so big 𝜁. Third term: cross-price impact on intermediate goods. Same as for second term. Fourth term: Labour income tax. Here, 𝜐𝑀𝑀 is big, so this term matters.

ECON 551: Lecture 12 31 of 55

The Double Dividend Debate

Putting a tax on something is effectively a tax on consumption. This increases the ‘real’ price of consumption, so implicitly acts as a tax on labour. (Bovenberg and de Mooij 1994)  At best, lowering taxes could totally counteract this impact.  But if you don’t exactly lower the right taxes, you’re not going to have your full dividend. Bovenberg and Goulder (2003 Handbook) use this terminology:  Revenue recycling effect: impact of lowering taxes  Tax interaction effect: cost of pollution tax interacting with other (e.g. labour ) taxes.

ECON 551: Lecture 12 32 of 55

ECON 551: Lecture 12 33 of 55

Agenda

• 1. Overview
• 2. Five methods of pollution abatement
• 3. Double Dividend Debate
• 4. Choice of Instruments
• 5. Canadian Policy Developments

ECON 551: Lecture 12 34 of 55

The Weitzman model

Should we focus on prices or quantities?  Weitzman (1974 REStud) argued it depended on uncertainty about the cost and benefit curves.  Abatement cost curve slopes up with amount of abatement.  Benefit curve slopes down with the amount of abatement.  Also depends on slopes of the two curves.  Let’s check out why…

ECON 551: Lecture 12 35 of 55

The Weitzman model

(p. 1529 Handbook chapter by Bovenberg and Goulder)  E is for expected; R is for realized.  Set tax and quantity at t* and a*, where expected MB=MC.  With cost uncertainty, tax leads to loss of triangle ABC; quantity regulation to loss of CDE.  Which is better depends on slopes of MC and MB.  With benefit uncertainty, we lose ABC either way.

ECON 551: Lecture 12 36 of 55

Other models

Dewees (1976 CJE)  Some say: impact of abatement subsidy and effluent tax the same.  But that only holds in very special conditions.

• Need high information about where to set base point for subsidy.
• Entry and exit from industry going to depend on this.
• Affects what part of cost curve you’re operating on.

ECON 551: Lecture 12 37 of 55

Comparing carbon taxes and cap and trade:

Goulder and Schein (2013) compare what we know about carbon taxes and cap and trade Four ways that pricing and cap and trade the same: i. incentives to reduce emissions: marginal incentives the same.

• ii. Compensation for distributional impacts: same options open to both.
• iii. Int’l competitiveness: border adjustments; mechanisms for subsidies to some

industiries.

• iv. Offsets: either can handle them.

ECON 551: Lecture 12 38 of 55

Comparing carbon taxes and cap and trade:

Goulder and Schein (2013): Ways they are different: I. Admin costs: either can be admin’d upstream or downstream. But for CnT, you need to monitor emissions and who has the permits.

• II. Emissions price volatility: a win for carbon tax.
• III. Addressing uncertainty: quantity uncertainty is advantage to cap n trade.

Weitzman model evidence: Recent evidence suggests cost function is steeper; this supports carbon tax.

• IV. Interactions with other policies. For cap n trade, this affects the price of the

permits, so you don’t get any additional benefit from other policies.

• V. Monopoly power by producers could capture rent under cap and trade.
• VI. Revenue recycling easier under carbon tax: given US institutions.
• VII. Int’l linkages: they argue carbon tax easier.
• VIII. Lobbying / coverage: under carbon tax, industries lobby to be excluded. Under

CnT, they lobby for free permits within the system.

• IX. Political perceptions: who knows.

ECON 551: Lecture 12 39 of 55

Agenda

• 1. Overview
• 2. Five methods of pollution abatement
• 3. Double Dividend Debate
• 4. Choice of Instruments
• 5. Canadian Policy Developments

ECON 551: Lecture 12 40 of 55

Pan-Canadian Framework on Clean Growth and Climate Change:

Agreement between federal/provs/territories in December 2016.  MB and SK didn’t agree….ON and NB pulled out…  http://www.cbc.ca/news/politics/trudeau-premiers-climate-deal- 1.3888244

Catherine McKenna, Minister of Environment and Climate Change 2015-2019.

Features:  Minimum $10/tonne in 2018  +$10 a year until reaching $50/tonne in 2022.  Either carbon tax or a cap and trade.  Provinces + Territories can use revenue as they like.  Feds will impose a price if province has no system: the backstop. Advantages:  Recognizes existing plans (e.g. BC).  Design can match diverse needs: not ‘one size fits all’.

 Revenues mostly stay in the province where taxes are raised.

ECON 551: Lecture 12 41 of 55

Federal backstop: Carbon tax rates

ECON 551: Lecture 12 42 of 55

Cap and trade example: Auction Results

Western Climate Intiative (California and Quebec): http://www.wci-inc.org/ https://www.ontario.ca/page/september-2017-auction-3-auction-summary-results-report

ECON 551: Lecture 12 43 of 55

Cap and trade example: Auction Results

ECON 551: Lecture 12 44 of 55

Cap and trade example: Auction Results

ECON 551: Lecture 12 45 of 55

Alberta: Output Based Allocations

https://www.alberta.ca/output-based-allocation-engagement.aspx Motivation: if other places don’t have carbon tax, exports put at disadvantage. Question: If firms get a cheque back for each barrel of oil produced, how does this lead to lower carbon use? (Hint: think about marginal production incentives…)

ECON 551: Lecture 12 46 of 55

Alberta: Carbon efficiency across sites

ECON 551: Lecture 12 47 of 55

Alberta: Will this work?

ECON 551: Lecture 12 48 of 55

Federal carbon price floor: All the links

https://www.canada.ca/en/environment-climate-change/services/climate-change/pricing-pollution- how-it-will-work.html

GoC Federal backstop https://www.canada.ca/en/services/environment/weather/climatechange/tech nical-paper-federal-carbon-pricing-backstop.html BC carbon tax https://www2.gov.bc.ca/gov/content/environment/climate-change/planning- and-action/carbon-tax AB carbon tax https://www.alberta.ca/climate-carbon-pricing.aspx SK Fed backstop https://globalnews.ca/news/3462367/sask-premier-describes-federal-carbon- tax-plan-as-a-ransom-note/ MB Fed backstop http://www.gov.mb.ca/climateandgreenplan/index.html ON Fed backstop https://www.ontario.ca/page/cap-and-trade (Pre Doug Ford) QC Cap and trade http://www.mddelcc.gouv.qc.ca/changements/carbone/Systeme- plafonnement-droits-GES-en.htm NB Fed backstop https://ipolitics.ca/2018/08/27/new-brunswick-votes-no-carbon-tax-tory- leader-says/ NS Cap and trade https://climatechange.novascotia.ca/proposed-cap-and-trade-program PE Carbon tax https://www.cbc.ca/news/canada/prince-edward-island/pei-carbon-pricing- reaction-1.4876738 NL Carbon tax https://www.releases.gov.nl.ca/releases/2018/mae/1023n01.aspx YT Carbon tax https://www.cbc.ca/news/canada/north/yukon-premier-carbon-tax-plan- 1.4885136 NT Carbon tax https://www.fin.gov.nt.ca/en/carbon-pricing NU Fed backstop https://www.canada.ca/en/environment-climate-change/services/climate- change/pricing-pollution-how-it-will-work/nunavut.html

ECON 551: Lecture 12 49 of 55

Three questions to ponder:

• 1. Will it matter?
• 2. Will it hurt business?
• 3. What happens with the revenues?

ECON 551: Lecture 12 50 of 55

Will carbon pricing matter?

Will we reach the targets we promised to the world in Paris?

Graphic courtesy Alastair Fraser, Ph.D. Source

ECON 551: Lecture 12 51 of 55

Will carbon pricing hurt business?

If we have a carbon tax and other countries don’t, will production of goods just move to countries that don’t price carbon?  If carbon emissions just get shifted, that doesn’t reduce global emissions. Solution 1:  Let exchange rates adjust and deal with it.  Canadian dollar becomes cheaper; restores our competitive advantage. Solution 2:  Transfer some revenue back to export-oriented business.  AB and Govt of Canada use “Output-Based Allocations” system.

ECON 551: Lecture 12 52 of 55

How to disperse carbon pricing revenues?

What to do with the revenues? It’s at the centre of the political debate. May revenue recycling options: i) Equal per capita household transfers (Federal backstop) ii) Low-income rebates (like GST cheque) (AB, BC) iii) Cut personal and corporate taxes (BC) iv) Spend on transit and other climate initiatives (BC 2017+; AB)

ECON 551: Lecture 12 53 of 55

Revenue Neutral?

Part 2 of BC’s Carbon Tax Act (repealed 2017…)

ECON 551: Lecture 12 54 of 55

Summary:

For emissions taxes, we have gone from a professor’s idea (Pigou) to application in about 100 years. On cap and trade, we have gone from a professor’s idea (Dales) to practical application in about 50 years. According to the Intergovernmental Panel on Climate Change, we are running out of time to act. Will carbon pricing work? We will see.  Just the future of the planet at stake—no pressure! 😊

ECON 551: Lecture 12 55 of 55

Agenda

• 1. Overview
• 2. Five methods of pollution abatement
• 3. Double Dividend Debate
• 4. Choice of Instruments
• 5. Canadian Policy Developments

…we are done!