AAEC 6524: Environmental Theory and Policy Analysis. Outline. Introduction to the Theory of Environmental Policy, Part A. Klaus Moeltner Spring 2017

Transcription

1 AAEC 6524: Environmental Theory and Policy Analysis to the Theory of Environmental Policy, Part A Klaus Moeltner Spring 2017 January 16, 2017 Outline

2 More realistic setup (many firms & households) Focus on interaction between firms (polluters) and households (victims) Game-theoretic perspective Role of transaction costs (cost of coordination), information, mitigating actions Polluter-pays principle (PPP) Overview of environmental policy instruments Setup j = 1... J firms coal-burning electrical plants, emit SO2: e j output price takers uniform contribution to total emission (non-spatial model): E = J e j j=1 i = 1... I households All exposed to same aggregate pollution level (non-spatial)

3 Damage function U i = y i D i (E) (1) quasi-linear in income: MUI=1, easy to express things in $ s D (E) > 0 D (E) 0 D (E) = I D i (E) i=1 aggregate (=social) damage function Abatement costs assume: cost of decreasing E ( abatement ) separable from other costs abatement cost function C j (e j ) C j (ê j ) = 0 C j (e j ) > 0, e j < ê j MAC j (e j ) = C j (e j) > 0, e j < ê j marginal abatement cost function MAC j (e j) = C j (e j ) 0, e j < ê j

4 Social planner s optimization problem Based on Social objective function: min e 1,e 2,...,e J SC (e 1, e 2,..., e J ) = J C j (e j ) + D (E) (2) j=1 J FOC s (differentiate w.r.t. each e j ): C j (e j) + D (E) = 0 j Efficiency conditions FOCs imply: C j (e j) = D (E) j Marginal cost of abatement (MAC) = marginal damage cost ( marginal benefit of abatement ) C j (e j) = C k (e k), k j MAC must be equal across all firms. Any policy that can achieve this is efficient

5 Graphical Analysis: Efficiency Figure 3.1: Pollution damages, costs, and efficient emission level C 1 (e 1 ) C1( e 1) D(E) D(E * ) 0 e 1 ê 0 1 e 1 E * E C 1 (e 1 ) C 2 (e 2 ) D (E) C ( e) 1 1 C ( e ) * 1 1 c b a * C2( e2) D (E * ) f C2 ( e2) e d g 0 e 1 * e 1 ê 1 e 1 0 * e 2 e 2 2 ê e 2 0 E * E Panel A Panel B Panel C Fixed some types 8 July 2011 Coase Coase (1960): Under some circumstances efficient outcome can be reached through negotiation alone No intervention needed. Another conceptual starting point. Highlights role of property rights

6 Setup Coase Simplify previous problem to 1 firm, 1 household. E = e 1 D (E) = D 1 (e) Sequential game (PRs with HH) Coase 1. Firm proposes contract (E, p) p=transfer payment 2. HH accepts or rejects transaction costs tr, borne by firm in this case assumption: complete and perfect information (preferences, technologies, payoffs)

7 Firm s problem Firm knows: HH accepts if p D (E) Coase Will offer p = D (E). Now find optimal E: min E TC (E) = C (E) + D (E) + tr (3) FOC: C (E) = D (E) Participation condition: C (E) + D (E) + tr C (0), or tr C (0) (C (E) + D (E)) transaction costs must not exceed social gains from negotiation. Sequential game (PR s with firm) Coase 1. HH proposes contract (E, p) p=transfer payment 2. firm accepts or rejects transaction costs tr, borne by HH in this case assumption: complete and perfect information (preferences, technologies, payoffs)

8 HH s problem Coase HH knows: firm accepts if p C (E) Will offer p = C (E). Now find optimal E: max E (y (C (E) + D (E) + tr)) (4) FOC: C (E) = D (E) Participation condition: C (E) + D (E) + tr D (ê 1 ), or tr D (ê 1 ) (C (E) + D (E)) transaction costs must not exceed social gains from negotiation. Coase Coase Coase theorem A imposes externality on B. If transaction costs are sufficiently small, irrespective of initial allocation of property rights: i) There exists an efficient contract with soc. opt. level of pollution E and transfer payment p. This leaves both parties at least as well off as before contract. ii) (added by Schweizer, 1987): Efficient contract can be interpreted as sub-game perfect equilibrium: a) non-pr owner suggests contract b) PR owner accepts

9 Coase Also known as Extended Polluter Responsibility (EPR). Emitter must bear both cost of prevention and pay for damages. Dates back to Sweden, 1975 In spririt adopted by many countries around the world U.S.: Implemented in some contexts (e.g. Superfund sites), but not (fully) in others (e.g. Clean Water Act) PPP and Coase Coase For our model: PPP consistent with HHs have property rights for emissions. But PPP ADDS payment of compensation to model. Won t change results - just interpret payment p as compensation for damages in ((3)) BUT: Efficiency outcome breaks down if we implement PPP AND HHs can mitigate against damages

10 Extended Social Planner Model Coase Suppose HH can purchase x at price p x to mitigate damages. New Damage function D (E, x), with D x < 0, D Ex < 0, Intuition: x decreases both total damages and marginal damages. D (E, x) + p x x = effective damage function Social planner s problem Coase min E,x (C (E) + D (E, x) + p xx) (5) C (E) = D E (E, x) D x (E, x) = p x (6) As before: Marginal abatement cost = marginal damage New: marg. benefit of private mitigation = marginal cost (price of x)

11 Coase Bargaining under PPP Coase HH gets fully compensated by firm no incentives to purchase x So firm will have to make payment p = D (E, 0) Firm s problem (assume tr = 0 for simplicity): min E TC (E) = C (E) + D (E, 0) (7) C (E) = D E (E, 0) (8) Comparison Coase Efficiency result: C (E) = D E (E, x) D x (E, x) = p x Implies: x > 0 Bargaining result: C (E) = D E (E, 0) HH underinvests in private mitigation. Inefficiently high level of abatement (low level of E)

12 E too low Coase D E (E, x) < D E (E, 0) since D E,x < 0, by assumption Also, by assumption: C (E) > 0 (marg. abatement costs increase with E) Therefore: E { C (E) = D E (E, x)} E { C (E) = D E (E, 0)}