Cost Benefit Analysis. April 15, 2018

Similar documents
Theoretical Tools of Public Finance. 131 Undergraduate Public Economics Emmanuel Saez UC Berkeley

Lecture 13. ECON 4910 Spring Monetary valuation The ethics and politics of cost benefit analysis

PARTIAL EQUILIBRIUM Welfare Analysis

Tax Incidence January 22, 2015

UTILITY THEORY AND WELFARE ECONOMICS

Optimal tax and transfer policy

Optimal Progressivity

Public Finance and Public Policy: Responsibilities and Limitations of Government. Presentation notes, chapter 9. Arye L. Hillman

Pareto Concepts 1 / 46

Pareto Concepts 1 / 46

1 Unemployment Insurance

Optimal Taxation : (c) Optimal Income Taxation

Benefit-Cost Analysis: Introduction and Overview

Lecture 5. Varian, Ch. 8; MWG, Chs. 3.E, 3.G, and 3.H. 1 Summary of Lectures 1, 2, and 3: Production theory and duality

Public Economics Taxation I: Incidence and E ciency Costs

10.11 CONCEPTUAL PROBLEMS IN MEASURING SURPLUS. Consumer surplus is an ad-hoc measure, not derived from a welfare measure

Econ 2450B, Topic 3: Commodities and Public Goods with Redistributive Concerns

Lecture 3 ( 3): April 20 and 22, 2004 Demand, Supply, and Price Stiglitz: pp

PARTIAL EQUILIBRIUM Welfare Analysis. Welfare Analysis. Pareto Efficiency. [See Chap 12]

Chapter 3 Introduction to the General Equilibrium and to Welfare Economics

Chapter 6: Correcting Market Distortions: Shadow Prices Wages & Discount Rates

Application: The Costs of Taxation

Optimal Labor Income Taxation. Thomas Piketty, Paris School of Economics Emmanuel Saez, UC Berkeley PE Handbook Conference, Berkeley December 2011

THE BOADWAY PARADOX REVISITED

THEORETICAL TOOLS OF PUBLIC FINANCE

ECON 340/ Zenginobuz Fall 2011 STUDY QUESTIONS FOR THE FINAL. x y z w u A u B

Gasoline Taxes and Externalities

Measuring Ex-Ante Welfare in Insurance Markets

Economics 230a, Fall 2014 Lecture Note 7: Externalities, the Marginal Cost of Public Funds, and Imperfect Competition

Econ 230B Spring FINAL EXAM: Solutions

Economics 11: Solutions to Practice Final

Economic evaluation of special events: reconciling economic impact and cost benefit analysis. Larry Dwyer

Chapter 3. A Consumer s Constrained Choice

Economics 2450A: Public Economics Section 1-2: Uncompensated and Compensated Elasticities; Static and Dynamic Labor Supply

Introduction to Economics I: Consumer Theory

PROBLEM SET 7 ANSWERS: Answers to Exercises in Jean Tirole s Theory of Industrial Organization

Introduction of Health Economics

Economics 101A (Lecture 24) Stefano DellaVigna

Mock Midterm 2B. t 1 + (t 4)(t + 1) = 5 = 5. 0 = lim. t 4 + (t 4)(t + 1) = 80

Lecture 12: Taxes. Session ID: DDEE. EC101 DD & EE / Manove Taxes & International Trade p 1. EC101 DD & EE / Manove Clicker Question p 2

Chapter 3: Model of Consumer Behavior

Recall the idea of diminishing marginal utility of income. Recall the discussion that utility functions are ordinal rather than cardinal.

University of Victoria. Economics 325 Public Economics SOLUTIONS

A Course in Environmental Economics: Theory, Policy, and Practice. Daniel J. Phaneuf and Till Requate

1 Excess burden of taxation

Welfare Economics. Jan Abrell Centre for Energy Policy and Economics (CEPE) D-MTEC, ETH Zurich. Welfare Economics

The Policy Elasticity

AAEC 6524: Environmental Economic Theory and Policy Analysis. Outline. Introduction to Non-Market Valuation Part A. Klaus Moeltner Spring 2017

MARKET-BASED APPROACH

Closed book/notes exam. No computer, calculator, or any electronic device allowed.

Econ 551 Government Finance: Revenues Winter 2018

ECON 4624 Income taxation 1/24

Lecture 13: Efficiency vs. Equity & International Trade

Introductory Economics of Taxation. Lecture 1: The definition of taxes, types of taxes and tax rules, types of progressivity of taxes

} Number of floors, presence of a garden, number of bedrooms, number of bathrooms, square footage of the house, type of house, age, materials, etc.

Solutions to Assignment #2

14.41 Final Exam Jonathan Gruber. True/False/Uncertain (95% of credit based on explanation; 5 minutes each)

Recitation #6 Week 02/15/2009 to 02/21/2009. Chapter 7 - Taxes

Generalized Social Marginal Welfare Weights for Optimal Tax Theory

Test 2 March 24, 2010 Chapters 25 and 5-9

MB (polluter) MC (pollutee) Water Pollution. Full pollution. Zero pollution

Revenue Equivalence and Income Taxation

Final Term Papers. Fall 2009 (Session 03a) ECO401. (Group is not responsible for any solved content) Subscribe to VU SMS Alert Service

Chapter 20: Cost Benefit Analysis

Economics 230a, Fall 2014 Lecture Note 9: Dynamic Taxation II Optimal Capital Taxation

Market failure Redistribution- Tax or subsidy Restrict or mandate private sale or purchase Public provision Public financing of private provision

Mock Examination 2010

Taxation and Efficiency : (a) : The Expenditure Function

Unit 4.3: Uncertainty

We will make several assumptions about these preferences:

Economics 2450A: Public Economics Section 7: Optimal Top Income Taxation

Measuring Ex-Ante Welfare in Insurance Markets

ECNS 432. Redemption Quiz/Final Exam Review (Answers)

Public Economics Lectures Part 1: Introduction

Income Inequality and Poverty (Chapter 20 in Mankiw & Taylor; reading Chapter 19 will also help)

Economics 101A (Lecture 25) Stefano DellaVigna

EconS Micro Theory I 1 Recitation #7 - Competitive Markets

Exam 1 Review Problems

The source of the leaky bucket is shown in figures 7.5a and 7.5b

ECON 6022B Problem Set 2 Suggested Solutions Fall 2011

PUBLIC FINANCE FOR PUBLIC GOODS

Advanced Risk Management

is a concept that relates the responsiveness (or sensitivity) of one variable to a change in another variable. Elasticity of A with respect to B = %

Module 10. Lecture 37

Lecture 8. Application: the cost of taxation

Public Finance: The Economics of Taxation. The Economics of Taxation. Taxes: Basic Concepts

The Collective Model of Household : Theory and Calibration of an Equilibrium Model

Micro Theory I Assignment #5 - Answer key

The theory of taxation/2 (ch. 19 Stiglitz, ch. 20 Gruber, ch.14 Rosen)) Taxation and economic efficiency

Competitive Markets. Market supply Competitive equilibrium Total surplus and efficiency Taxes and subsidies Price maintenance Application: Imports

Public Economics. Lecture 4: Public goods and externalities. Marc Sangnier , Spring semester

Finance: Risk Management

Theory of Consumer Behavior First, we need to define the agents' goals and limitations (if any) in their ability to achieve those goals.

A note on Cost Benefit Analysis, the Marginal Cost of Public Funds, and the Marginal Excess Burden of Taxes

Notes on Macroeconomic Theory. Steve Williamson Dept. of Economics Washington University in St. Louis St. Louis, MO 63130

A simple proof of the efficiency of the poll tax

Chapter Four. Utility Functions. Utility Functions. Utility Functions. Utility

10. Fiscal Policy and the Government Budget

Application of Welfare Analysis: The Costs of Taxation

INDIVIDUAL AND HOUSEHOLD WILLINGNESS TO PAY FOR PUBLIC GOODS JOHN QUIGGIN

Transcription:

Cost Benefit Analysis April 15, 2018

Comparing the social value of different policy projects Policy makers can only implement a limited number of projects. n order to implement those with highest social value one needs a common metric, a goodness measure, that can be aggregated to social value. Social value should be based on individualism: the costs and benefits of a policy j on individual i should be based person i 0 s own judgement. pareto: ifpolicyj makes at least one person better off and no one worse off (according to their own preferences) then implementing j increases goodness or SW. welfarism: socialorderingofpoliciesmustbebased exclusively on, and taken as an increasing function of, the respective collections of individual utilities of these policies.

Comparing the social value of policy projects CBA based public policy is extremely ambitious 1. Assess the predicted effects of a policy (should be based on estimates of the causal consequences of a policy) 2. Assess how individuals value the policy consequences (assign monetary value to the consequences; find the right prices) 3. Compare losses and benefits across different individuals (Pareto is a no-brainer) (Find the right weights to aggregate individual values to asocialvalue).

An Example: Environmental speed limit in Oslo Since 2004 the speed limit at certain roads in and around Oslo has been reduced from 80 to 60 during the winter months (from 1. November) to increase air quality. s this a good policy? CBA Effects Prices Social value How does the intervention affect: air quality, traveling time, accidents,... What is better air quality worth for individuals, what is reduced traveling time worth, fewer accidents How do we weigh the lost value for those who get longer traveling time against the value for those who enjoy better air quality

Plan for the lectures n CBA there is not much discussion of how difficult it is to predict the effects of a policy change. t is hard to estimate policy impacts. You need a model and a clever design to estimate the parameters of that model (compare potential outcomes (with and without a policy). CBA centers around how to get the prices correct; when should we adjust market prices when they exist, how should we value things that are not traded (what is the value of a life, what is the value of saving ice-bears) how should we discount the future, and has should we take account of uncertainty. Using the right prices is important in order to assess the appropriate costs and benefits at the individual level, but the weighing function that aggregates individual values to an overall social value is also important. And the shortcuts done there often hide many important value judgments. 1. Social Welfare 2. Getting the prices right

Comparing individual values: Social welfare functions Estimating the effects of a policy (whether or not environmental quality, or time to travel, is altered by an environmental speed limit) and assigning how individuals value these changes are (in principle) empirical questions. Questions that can be answered with data. How we should weigh individual values to arrive at the social value of a policy or a project, is a normative question. This question cannot be answered with data (facts). You need good arguments. individualism and welfarism =)Social welfare W = w(u 1,...u N ) with @W @u i > 0.

Social Welfare The shape of the SW function reflects value judgements, such as the concern for equality. f social welfare is parametrized by the iso-elastic welfare function W = NX ui 1 i=1, will capture the concern for equality. The higher the more concave is the welfare function. 1 = 0 =) utilitarianism: unweighted sum of utilities = 0 =) rawlsian: all weight on the worst off. individual utility is not observable. The shape of u that is used in CBA is also based on a value judgment. A key issue is to what extent marginal utility of income decreases in income.

Monetary representation of individual utility (monetary representation of policy reforms) One way to get a monetary representation of utility is to ask for the minimum income i needs to attain a certain utility y i = e(p r, ū) :min (p r x u(x) > ū) that gives u(y i ) and u 00 captures how marginal utility of income declines in income. f we assume iso-elastic utility of income u = y 1 1 we can write social welfare in terms of the distribution of real income in the society:, with = + + w = NX i=1 f both and are large numbers, the social welfare effect of a benefit to a very rich person is small because (i) the marginal utility of income is decreasing at the personal level and (ii) marginal social welfare decreases in the income level. 1 yi 1

Monetary representation of individual utility (allows monetary evaluation of policy reforms) Most CBA disregards distributional issues; assumes = 0 =)evaluates projects according to aggregate individual real income effects (does not take into account who gets the benefits and costs). Why? don t want to make value judgement (but assuming judgment. = 0) is also a value separate efficiency and redistribution (implement policies that maximizes aggregate income, redistribute through tax policy) if a project A generates a change in aggregate income that is positive P 4yi A > 0 this implies that the winners can in principle compensate the losers. many policy projects, sometimes you win and sometimes you loose, evens out. none of these arguments are convincing. But it is of course very ambitious to find individual effects of policy projects. An alternative is to aggregate to groups (introducing a tax on suv cars will affect the buyers (4y B ) producers (4y S ) and the government (4y Gov ). 4w = r b 4y B + r S 4y S + r Gov 4y Gov.

Monetary representation of individual utility (allows monetary evaluation of policy reforms) How do we find 4y B? We can use the expenditure function. Suppose all other prices (p) (except for suv) are constant, while the after tax price for suv increases from p 1 suv > p 0 suv. e(p, p 1 suv, u 0 ) is then the income suv drivers need in order to reach the welfare they had before the suv tax was introduced.4y B (t suv )= e(p, p 1 suv, u 0 ) y 0. Can also use this logic to assess the effect of forbidding suv cars. 4y B (suv forbidden) =e(p, p 1 suv!1, u 0 ) y 0 Willingness to pay: Basically asking suv drivers what they would be willing to pay in order to get rid of the policy that forbids suvs. Note that if suv drivers are very rich they are willing to pay a lot to be able to drive cars with big engines (because marginal utility of income is low for them). f CBA disregards distributional concerns ( = 0)=)the stated loss of the ban on suvs will be very high. Willingness to pay without weights =)systematic priority of the interests of the rich. Social welfare is given by w(u 1, u 2,...u N ) if the real income effect on i and j is dy i and dy j the social welfare effect @u

CBA and the private value of a project Suppose we want to asses the social costs and benefits of a project A, why does it differ from the commercial value of the project (if a firm implemented A). Or; why cant we use market prices to assess the social costs and benefits of a project. 1. Market prices do not reflect the true marginal costs and benefits 1.1 mperfect competition 1.2 externalities 1.3 tax distortions 1.4 information asymmetries 2. There are no market price 2.1 public goods; basic research... 2.2 travel time 2.3 clean air 2.4 saved lives 3. Projects may require public funds; marginal costs of public funds 4. Effects over many periods and the social discount rate may differ from private discount rate. 5. Uncertainty.

The decision rule Suppose B P t and C P t are the true monetary benefits and costs of a project in period t. P a has a positive net present value of NPV a = 1X t=0 B t C t (1 + r) t where r is the correct discount rate to use to compare social value across time. The decision rule is simple: f there is no constrain on the number of project that can be implemented, carry out all with NPV > 0. f there is a constraint on number of projects, carry out those with the highest NPV

True benefits and costs require correct prices P may produce goods and services and use resources that are traded in the market or effects that are not traded - where there are no market price. How should we price the effects. The market price is often distorted and if the good is not traded there is no price =)must calculate the true value of what is traded, find the shadow price (the increase in social value if one unit of this good was injected into the economy)

True benefits and costs require correct prices Example Project P needs 1000 workers for one year. Suppose the project draw workers from three sources. Afractionf e worked elsewhere. They earned w(1 t) per unit of labour (w is before tax salary and t is the tax rate) Afractionf v where voluntary unemployed (looking for jobs) Afractionf iv where involuntary unemployed. What is the true labour costs of this project?

Social value and shadow prices More general point: resources required in a public project (dz > 0) may come from reduced consumption (dx < 0) or increased production (dy > 0). f there are tax distortions the alternative value, the shadow price, of reduced consumption and increased production are different. Suppose a public project needs one unit of good k. Supposetheconsumer price for this good in the market is p k. The producers get p k t k (there is a tax distortion). f the public project reduces consumption the shadow price is p k if the project does not reduce consumption but increases production the alternative value - shadow price - is p k t k. n addition a change in price of good k may have effects in another markets j 6= 0. f there are no distortions there there is no first order @x j @z k effect in social value even if @x j @z k 6= 0. But if there are tax distortions (or other distortions) in market j the shadow costs must also include these @x indirect effects t j j @z k.

Shadow price of government revenue f a project is not self financed but needs to be financed through tax revenue we need to take account of the excess burden of taxation and the fact that the marginal cost of increasing government revenue with 1 NOK is equal to 1 +, where is the excess burden of taxation.

Excess burden A tax affects behavior (what we consume, how much we work, how much we save..) through two channels; by altering relative prices (sub-eff) and by reducing the purchasing power of private income and wealth (inc-eff). t is only the sub-eff that creates an efficiency loss. To measure EB we need to get rid of the income effects an example with quasi linear utility work with Hicksian (compensated) demand functions

EB in a simple case with quasi linear utility Two goods x and y. u(x, y) =h(x) +y with h(x) increasing and concave Per unit tax on good x, thetaxinclusivepriceisq = p + y is untaxed and the price is normalized to 1. Consumers maximize u(x, y) s.t. y +(p + )x = z Demand function implicitly defined by h 0 (x) =(p + ) Producers use c(s) units of good y to produce s units of x; c 0 > 0, c 00 > 0 profits = ps " S = @S p @p S c(s) f.o.c. p = c 0 (s(p)) with elasticity

EB in a simple case with quasi linear utility illustration of the excess burden of taxation

EB in a simple case with quasilinear utility measure excess burden EB = 1 2 d dq dq = S 0 (p)dp dp = D S D dt to get =)EB = 1 D S 2 S D pq( d p )2 Triangle ) EB! 0ifd! 0. There is no first order EB effect of a very very small tax starting from a zero tax (the economic value of the transaction the tax preempted is 0)

Two important lessons EB increases with the square of the excise tax (more than proportional), argument for a broad tax base EB increases in the drop in demand caused by the tax, and this drop depend on the demand and supply elasticity. Higher taxes on inelastic tax bases. =)Marginal cost of public funds depend on the tax level in the economy and on the structure of the tax rates.

How to value things that do not have a price Same principle: Apply individuals willingness to pay (wtp) (or willingness to accept). Practical problems since individuals have not revealed willingness to pay in a market. Sometimes possible to estimate the willingness to pay based on choice data. Can use the market indirectly to assess the value the value of reduced local pollution will show up in local house prices less congestion, reduced travel time - can use wages to assess this effect. lower risk of death - ex ante willingness to pay for this reduction. Can use the wage difference in two jobs that are equal, except a difference in risk of injuries/death. Hedonic prices

Contingent Valuation Survey-based valuation technique Survey representative sample about their WTP or their WTA for environmental goods and services. Applications: Benefits of improving air/water quality Benefits of wilderness areas Benefits of outdoor recreation opportunities

Risk f the benefits from a project are uncertain the expected benefits from the project will not reflect the true social value if those affected by the project are risk averse. The value of an uncertain project will decline in risk aversion and in the risk of the project. Difficult to calculate the loss related to risk; need to make assumption about the concavity of the utility function The government has many project, and if returns are to some extent independent the portfolio is diversified and the risk of one project can be ignored.

Discount rate What is the correct rate to discount future social benefits and costs? The principle is straightforward: use the discount rate used by households to compare current and future consumption. n a competitive market where savings do not have externalities could use the interest rate on savings f there are taxes on capital income, the appropriate consumption discount rate would be the after-tax return that households obtain on their savings (or pay on their borrowing). Externalities, liquidity constraints,... complicates things Hyperbolic discounting A big theme is intergenerational comparisons - to what extent should we take into account the welfare of unborn generations.