1 Economie Publique II February-May 2010 Prof. A. Estache Lecture 1 Overview

1

Economie Publique IIFebruary-May 2010

Prof. A. Estache

Lecture 1Overview

2

Some organizational issues

• Course organization

• Course content and coverage

• What I assume you know already…

• Required readings and what they mean to you

• Exam…a first sense of what you will get

• What this course could do for you

3

Course organization• Lectures are on Wednesdays 10-12 at H5.164• You will have 12 lectures• I will post the lectures and any related material on

my web site (bottom of the entry page for a list of courses) http://164.15.69.62/index.php?option=com_comprofiler&task=userProfile&user=107&Itemid=263

• You will have to follow the reading list• Trying to read ahead of time will help you

understand the lectures• I will also post any readings not easily available in

pdf format on the course web site. • Always happy to meet with you or hear from you if

something is not working out the way you want…my email is [email protected]

http://164.15.69.62/index.php?option=com_comprofiler&task=userProfile&user=107&Itemid=263

http://164.15.69.62/index.php?option=com_comprofiler&task=userProfile&user=107&Itemid=263

4

Course content (1)

• This course will focus on how best to regulate firms with significant market power.– A crucial role of governments and hence

central to public economics knowledge

• It will get both into the theory and the practice of regulation.

• Network industries will offer the main illustration of the concepts

5

Course content (2)• Week 1: Introduction and Overview (…today!)• Week 2: Regulation with complete information• Week 3: Regulation with asymmetric information• Week 4: Extensions to the basic model• Week 5: The regulation of prices• Week 6: The regulation of quality• Week 7: Equity-efficiency trade-offs in regulation • Week 8: Incentives issues built-in the design of regulatory

institutions• Week 9: Financial Modeling of Regulated Industries• Week 10: Asset Valuation of Regulated Industries• Week 11: Cost of Capital of Regulated Industries• Week 12: Efficiency Measurement in Regulated Industries

6

What I assume you know already• You NEED to be familiar with everything you

have been taught on the theory of monopolies!!!!– If you don’t remember much, review it before next

class• IT IS CRITICAL since that is at the core of everything you will

see in this class

– To make it easier for you to review this theory, I have attached in the annex the slides from Varian on this topic…you really need to know this!

• You need to be able to set up an optimization problem under constraint

• You need to be able to have a sense of synthesis since you will have to read a fair amount of material

7

You need to remember everything that goes around this graph!

30

E ATC

80

$120

DMR

MC

Number of Kwh consumed

Euros per Kwh

Profits from standard monopoly profit maximization

8

MC

MR

Demand

Price Regulation: Government can make an outcome more efficient as long as PC < PR < PM

Competitive EquilibriumCompetitive Equilibrium

Monopolistic Monopolistic equilibriumequilibriumPM

PC

PR

9

Required readings (1)• 4 basic texts

– Armstrong and Sappington (2007), “Recent developments in the theory of regulation”, in Handbook of Industrial Organization (vol 3), edited by Armstrong and Porter, North Holland

– Estache, A. and M. Fay (2010) “Current debates on infrastructure policy“ in D.Leipziger (2010), Globalization, Growth Commission, Washington, D.C

– Estache, A. and L. Wren-Lewis (2009), “Towards a Theory of Regulation for Developing Countries: Following Jean-Jacques Laffont’s Lead”, (Forthcoming), Journal of Economic Literature (forthcoming)

– Estache, A., M. Rodriguez-Pardina, J.M. Rodriguez and G. Sember (2002), “An introduction to Financial and Economic Modeling for Utilities Regulators”, Policy Research Working Paper 3001, World Bank, Washington, D.C.

• Some additional required reading for some of the lectures

10

Required Reading (2)• For every required reading, I will

give you a few questions you need to be able to answer

• You will have to provide me with a summary of your replies. – These replies will be due 1 week

before the end of the course.– They should force you to make sure

you understand the material you are supposed to be familiar with.

11

Exam• You will also have to know well the lecture slides

– It will really help you to also know the required readings for the exam

– You will find some of the suggested reading useful if you have not understood the lectures in class.

• But these are not required and I will not ask any questions on those readings.

• They may simply be useful of you decide to specialize in regulation.

• Your understanding of the course material will be assessed by means of a three hour examination

• More details on the exam will be provided at the end of the course.

12

What this course could do for you

• Most courses on regulation of network industries are highly technical…(e.g. Toulouse, Oxford, Florida, MIT, …)

• This course is a combination of some demanding theory and practice

• It will provide you with enough theory to give you a sense of what to deal with in a PhD thesis or research in the field of regulation

• It will provide you also with a set of tools you would need to have if you were to work as a consultant or in a regulatory agency or… in a regulated company!

13

Overview of the course• A working definition of regulation

• The policy problem

• The view of the world of regulation today

• The evolution of research

• A zoom on the lessons from theory on institutional design

• All the main dimensions of the measurement of the effects of regulation of network industries to help you in the real world

• Concluding comments

14

What exactly is regulation?• Government regulation of industry is

– local, federal or state government control – of individual or firm behavior – via 3 main instruments

• Prices (level and structure)• quantity (of service, environmental, safety,…)• quality of goods and services produced.

– E.g. setting rates for electricity service. – E.g. imposing service obligations such as the

number of stops on a bus or rail itinerary– E.g. setting emission standards for cars

15

The dimensions of the regulation problem

• The actors

• Their objectives

• Their constraints

• The trade-offs

16

The actors

• the government (local, national)• a firm (monopolist)• the users (final or intermediate)• a regulation and a competition agency• the tax payers• other firms (as users or competitors)• other agencies• ...

17

The objectives of the actors• the monopolist maximizes its private profits• the gvt wants to foster efficiency and fairly distribute

the rents: – accepts monopoly structure (size economies) and maximizes

social welfare…but private agenda (re-election,…)

• the users maximize their net utility• the agency is a surrogate for the government…but

private agenda (maximizes its net resources risks of capture). Governance problems.

• the tax payers react to the fiscal burden that distorts relative prices

• the other firms maximize their profits horizontal dimension of the strategy

• other agencies: idem

18

The constraints on policymakers• technological and economic

– costs, preferences

• legal– privatization law, concession law, sector laws, antitrust law,…– contract design (including duration)– accounts rules and reporting obligation

• institutional– who is involved and who decides and how (price control, wage

control, barriers to entry)– enforcement power– financing opportunities

• informational– imperfect but symmetric information– asymmetric information: adverse selection, moral hazard

19

The trade-offs as the main dimension of the policy problems

• Efficiency: – concern of sector people/economists

• cost reflecting prices• cost minimizing production choices• getting the investments going as needed

• Fiscal and financial viewpoints:– concern of the finance team

• «fiscal pay-off to gvt » (annually/1 time shot)– Whether it goes to the treasury or to finance political campaigns and other expenses

by the politician is also relevant

• Social Concerns/Equity:– concerns of politicians

• lowest possible price and related considerations• best possible quality• largest cheap expensive looking investments

• Votes!!!– Concerns of the politicians

• Only real instrument consumers and taxpayers have to voice their happiness or discontent

• Governance– To ensure that accountability of all actors can be enforced

20

The big picture on the problem• Economists and Politicians are against

monopolists and agree they need regulation…but for different reasons:– economists: monopolists do not produce

enough– politicians: monopolists charge too much– …and the solution to their concerns could

converge: lower price means more volume– ..but often don ’t:

• economist wants price discrimination; politician prefers price cap which yields rationing but a politically manageable price

• economist wants to push competition (entry), politician not necessarily (strong emotions attached to ownership of assets supporting public services)

21

What has theory had to tell us about this policy problem?

• Traditional theory (with benevolent gvt and focus on market failures--monopolies, externalities, …)

• The recognition of the non-benevolence of government

• The information asymmetry issue as a revolutionary shock

• A zoom on the empirical research

22

Traditional theory• Assumes benevolent gvt

• i.e. assumes politicians have no private agenda!

• Focus on market failures– Solved by regulation or otherwise

• Pigouvian taxes & subsidies (1920s)• Marginal cost pricing debate (Boiteux (1949, 1956)) • Ramsey, peak load & other fancy pricing

(Baumol/Bradford (1970))• Coase theorem and bargaining (1960)• Demsetz auction and competition for the mkt (1968)

• Contestable markets (Baumol, Panzar, Willig (1970-80s))

23

The problems with traditional theory

–Problems are that regulation is:• Bounded by transaction costs

– Williamson (1960s), Dixit, Martimort (1990s)

• Bounded by information and related incentive problems

– Baron and Meyerson (1981) Laffont, Tirole, Armstrong, Milgrom, Newbery, Sappington, Vickers, Willig (1980-90s), …

• Often responsive (endogeneity issue)

• Driven by complex gvt motivation and structures and multiple gvt objectives

24

New theories: NON BENEVOLENT GVT (1)

• Capture theories• Role of auction design in achieving

competitive outcomes in regulated markets• Scope for renegotiation• Heterogeneity of goals from heterogeneous

gvts (multiprincipals) • Concerns for coordination• Political entrepreneurs vs. public service

concern• Governance aspects and voting

mechanisms

25


The various focuses of research

– Regulatory design• P cap, RoR, Hybrid regime

– Incentive mechanism design• Accounting explicitly for effort level and degree of information

asymmetry and related strategic concerns

– Institutional design• Degree of financial and political autonomy• Degree of corruption • Degree of competence• Governance structures

– Timing and sequencing of regulatory reforms

26


• Instruments– Agency models, game theory, advanced simulation

techniques

• Literature– Theory: Posner (1971), Stigler (1971), Peltzman (1976), Becker

(1983, 1985), more recently, Laffont-Tirole, Noll, Spiller, …

– Empirics: a lot on some aspects but little on many others!

• Problems:– Disagreements on degree of common knowledge of

information needed for regulation and related strategic interactions

– Bridge from theory to practice cuts lots of theory corners

27


Main practical messages are:• regulation and competition policy generate rents

worth fighting for …but they distort outcomes and users usually lose if decision makers are unaccountable

• there are trade-offs between rents and efforts level by the operators and short term losses by users may become long term losses if weak “regulators”

• Various types of risks matter to effort and investment incentives and should drive choices

• Distributional consequences can be dramatic

28

A zoom on what theory tells us about institutional design

• not much advise from traditional theory• many more possibilities offered by information

economics and agency theory on organizations• a lot more coming from recent developments on

collusion theory• need to build a positive theory of government

around these recent developments• need to recognize transaction costs and political

constraints

29

What are the problems leading to transaction costs?

• information asymmetries: demand elasticity? exact technologies? costs?

• limits to the markets as an implementation tool for policies private police? private regulators?

• incompleteness of contracts: non-contingent contracts + contracts viewed as non-binding by future gvts which renders renegotiation of some type unavoidable

• multiple principals: multiple government agencies share responsibility but fail to coordinate because discretionary politics drives decision-making by bureaucrats; all this reduces accountability of regulators

30

Why should we care? Because it tells us that

• structures matter:– distribution of regulatory rights – definition of regulatory goals – voting procedures driving decisions

• processes matter– timing of government intervention– length and span of control– design of communication channels

• Both affect regulatory outcomes: work on the efficiency-rent trade-off

31

The specific government failure to address

• lack of commitment of government and need for renegotiation

• multiprincipal nature of government

• discretion of political principal• discretion of regulator

32

Lack of Commitment and Unavoidable Renegotiation (1)

• new information can change the goals of the game

• degree of commitment of actors as well– under full commitment: optimal regulation is well

understood– without full commitment: strategic behavior by both

the gvt and the firm is likely and time inconsistency problem in the firm’s behavior, reducing incentive to be efficient in 1st period

• MAIN MESSAGE: Incentive compatibility constraints are hardened by simple existence of renegotiation

33

How to improve commitment?

• maintain flexibility to renegotiate but spell out renegotiation rules early on

• there is an optimal degree of separation of powers but this acts on incentives

• consider a sequential move of regulatory decisions and overincentivate in period 1

• address explicitly the speed of reform

• spell out checks and balances

• independence and accountability of agency

34

The Multiprincipal Nature of Government (1)

• multiple agencies dealing with regulation (environ, health, economic)

• it means multiple not necessarily cooperative objectives

• this means allocative distortions because coordination problems

• suboptimal decentralization cuts incentives; too many players decreased efficiency of regulatory decisions; risk of free riding

35

The Multiprincipal Nature of Government (2)

• if regulation of complementary activities, risk of overregulation and cuts incentives built in contracts due to lack of coordination

• if regulation of substitute activities, risk of underregulation (too much incentives) due to regulatory competition

36

Excessive discretion of the political and regulatory principal

• distinction between formal and real authority is crucial

• real authority in the hands of implementing agents (regulators/bureaucrats)

• formal authority in the hands of political decision makers

37

Need to reduce discretion of the Political Principal

• structures drive the distribution of power among interest groups

• politicians end to favor median voter (his constituency) over majority of voters (society as a whole)

• apparently suboptimal contracts (average cost pricing) now could have a rational: they tie the hands of politicians by cutting discretion

38

Need to reduce discretion of the Regulators

• increase accountability to deal with the imperfect monitoring of the regulators

• collusion proof constraint can reduce conflicts between legislative and regulators: low powered incentive contracts are usually better since cuts discretion (influences efficiency-rent trade-offs)

• need to avoid side contracts between regulators and interest groups: this requires more established, clear and transparent structures

39

How much empirical evidence do we get from academic research?

• It measures the effects of regulation and competition on– Average price levels– Price structures– Access pricing– Static efficiency

• Input distortions (Averch-Johnson effect (1962), Unions,…)• “X-inefficiency” (are we on the production frontier?)• Rent size• Cost of capital• Regulatory costs

– Dynamic efficiency• Rate of innovations and productivity growth

– Product quality and variety– Contract sustainability– Macroeconomic variables– Distributional implications

• Redistribution of existing rent between players• Allocation of rents from changes

40

The methodology of measurement (1)

• (1) Comparing regulated and unregulated firms and markets

• ( 2) Tracking variations in regulation intensity

• (3) Controlled environment experiments

• (4) Structural models and simulation

• [(5) Detailed case studies with statistically weak evidence]

41

The methodology of measurement (2)• (1) Comparing regulated and unregulated

firms and markets– Approaches

• Across countries, states (cross-section)• Historical data (time-series)

– Product/service markets– Financial markets (track the stock market)

– Additional use• Generates data for yardstick competition• Generates data for tariff revisions/price caps,…

– Problems:• Heterogeneity of regulatory regimes and enforcement• Endogeneity of regulation (responds to local conditions)• Specific legal environment and instruments• Timing, sequencing

42

The methodology of measurement (3)• (2) Tracking variations in regulation

intensity– Quantitative differences in regulatory

constraints• e.g. Different levels of rate of return allowed

– Differences in regulatory resources• e.g. budgets, skills, autonomy

– Specificity of instruments and procedures• e.g. anglo-saxon vs. napoleonian traditions

– Need for independent rating of quality of regulator

– Interactions with economic environment (inflation, growth, corruption, risks…)

43

The methodology of measurement (4)• (3) Controlled Environment Experiments

– Field experiments• Peak load pricing vs uniform• Bulk markets vs retail markets

– Problems: expensive to implement and hard to control correctly

– Laboratory experiments• Experimental economics

– Problems: risky extrapolations

44


• (4) Structural models and simulations

– (i) Demand and cost systems• Need to estimate cost and production functions• Need to estimate demand• Useful to simulate alternative regimes• A lot of good results available and used• Drawbacks:

– Endogeneity of input prices– Short term vs long term – Regulatory distortion– Changing environment – Interaction with environment

45


• (ii) Computable General Equilibrium Models– Solve models as if full price control by

monopoly or full competition and the comparison yields the size of the rent and hence the value of regulation:

• 0.3% of GDP for Argentina

• (iii) Counterfactuals– Reconstruct operators accounts and impose

new restrictions• Galal, et alt. : huge gains from privatization• Gasmi, Laffont, Sharkey: measures impact of

incentive based regimes vs cost + on telecom prices

46

Summing up the academic view of regulation theory:

• Difficult to come up with a single clean story due to heterogeneity of:– Initial conditions– Variables monitored– Dimension of regulation– Sectoral diversity

• Regulators and competition agencies have not generated the data that will help research make them better regulators, collectively at least

• But plenty of actions in the real world and useful to know how regulation sometimes work in practice

• So how do you go from theory to practice?

47

Back to basics: How to come up with fair regulation of a Natural Monopoly

B$15

$29

A

C

MC

$60

LRATC

50,000

DMR

85,000

100,000 Number of Household

s Served

Dollars

Unregulated monopoly

"Fair rate of return" production

F

48

This faire regulation builds on basic economics!

Average TARIFF

SUPPLY

DEMAND

Cost Recovery =

Allowed Revenue

Economic Signal Tariff Structure

Operational expenses or

OPEX

Capital expenses or

CAPEX

©World Bank Institute

49

A checklist of tools to do the job!• Asset valuation• Estimates of the cost of capital• Regulatory Accounting Guidelines• Tariff Rules• Penalties• Efficiency measures• The contract with the government• The economic and financial regulatory model• Indicators of financial viability• A clear view of the analytical decision criteria

50

The Tools of the RegulatorsAsset Valuation

• At the core of everything• With the cost of capital, allows the

determination of the allowed revenue for required efficiency levels

• With the specified deprecition rules, defines the amortization expenses allowed in OPEX forecast

• Be clear on which method you pick• Update regularly

51

The Tools of the RegulatorsEstimate of the cost of capital

• Minimum rate of return allowed to the operator of a monopoly

• Imposes going through a fair assessments of risks

• Require a methodological discussion

• To be reviewed periodically

52

The Tools of the RegulatorsRegulatory Accounting Guidelines• Core instrument to generate the information

required to regulate• Includes ALL guidelines

– Depreciation rules– Cost allocation rules– Expenditures forecast processes

• Unavoidable as few countries tend to generate good economic information for regulators

• Should be included in contract with operators

53

The Tools of the RegulatorsTariff rules

• Sector law and/or related decrees• Legal specification of tariff structure• Indexation rules• « pass-through » rules• Ordinary and extraordinary tariff revision

rules• Clarify behavior subject to penalties and

relevance for tariffs• Specify penalties in legal documents

54

The Tools of the RegulatorsEfficiency Measures

• Efficiency gain is economic rent to be redistributed => MUST be measured

• Announce how it will be measured

• Ensure accounting guidelines will generate the data needed

• Explain how results will be used in regulatory decisions

• Specify a related consultation process

55

The Tools of the RegulatorsThe contract with the government!

• Specifies rights and obligations of the various actors – Investments– Quality– Timing of obligations– Payment to gvt– Subsidies

• Includes information obligations

• Core reference document of regulator

56

The Tools of the RegulatorsThe financial and economic model

• Analytical framework which– Ensures internal consistency of:

• All obligation to all parties and any change to these • The combination of regulatory instruments,

• Assumptions on the behavior of actors

– Quantifies impact of options to all actors in case of conflict based on transparent rules

• Calculates the business’ s Internal Rate of Return– Which means nothing unless compared to the

cost of capital !!!

57

The Tools of the RegulatorsA clear decision criteria

• IF IRR = CoC => ideal– NPV of activity is 0

– No « excess » profit for the operator

– Comparable to a competitive markets returns

• If IRR > CoC => « Excessive » Profit=>Regulatory action needed t revision time (e.g. cut tariff, subsidies,…)

• If CoC > IRR => Risk of bankruptcy =>Regulatory action may be needed to ensiure service continuity (e.g.e:

subsidy, change in service obligations level or timing, …)

58

The Tools of the RegulatorsFinancial viability criteria

• Control of standard financial indicators: • What credit rating agencies would look at to check credit

worthiness

• Cash flows in term of equity, assets and free cash flows

• Debt ratios, liquidity, debt composition

• …also need to check the consequences of limited access to capital markets

ANNEX:

What Varian taught you about

Monopolies….

Pure Monopoly• A monopolized market has a single

seller.

• The monopolist’s demand curve is the (downward sloping) market demand curve.

• So the monopolist can alter the market price by adjusting its output level.

Pure Monopoly

Output Level, y

$/output unit

p(y)Higher output y causes alower market price, p(y).

Why Monopolies?

• What causes monopolies?– a legal fiat; e.g. US Postal Service

Why Monopolies?

• What causes monopolies?– a legal fiat; e.g. US Postal Service– a patent; e.g. a new drug

Why Monopolies?

• What causes monopolies?– a legal fiat; e.g. US Postal Service– a patent; e.g. a new drug– sole ownership of a resource; e.g. a toll

highway

Why Monopolies?


highway– formation of a cartel; e.g. OPEC

Why Monopolies?


highway– formation of a cartel; e.g. OPEC– large economies of scale; e.g. local utility

companies.

Pure Monopoly

• Suppose that the monopolist seeks to maximize its economic profit,

• What output level y* maximizes profit?

( ) ( ) ( ).y p y y c y

Profit-Maximization

( ) ( ) ( ).y p y y c y At the profit-maximizing output level y*

d ydy

ddy

p y ydc y

dy( )

( )( ) 0

so, for y = y*,

ddy

p y ydc y

dy( )

( ).

y

$R(y) = p(y)y

Profit-Maximization

$R(y) = p(y)y

c(y)

Profit-Maximization

y

Profit-Maximization

$R(y) = p(y)y

c(y)

y

(y)

Profit-Maximization

$R(y) = p(y)y

c(y)

y

(y)

y*

Profit-Maximization

$R(y) = p(y)y

c(y)

y

(y)

y*

Profit-Maximization

$R(y) = p(y)y

c(y)

y

(y)

y*

Profit-Maximization

$R(y) = p(y)y

c(y)

y

(y)

y*

At the profit-maximizingoutput level the slopes ofthe revenue and total costcurves are equal; MR(y*) = MC(y*).

Marginal RevenueMarginal revenue is the rate-of-change of revenue as the output level y increases;

MR yd

dyp y y p y y

dp ydy

( ) ( ) ( )( ).

Marginal RevenueMarginal revenue is the rate-of-change of revenue as the output level y increases;

MR yd

dyp y y p y y

dp ydy

( ) ( ) ( )( ).

dp(y)/dy is the slope of the market inversedemand function so dp(y)/dy < 0. Therefore

MR y p y ydp y

dyp y( ) ( )

( )( )

for y > 0.

Marginal RevenueE.g. if p(y) = a - by then R(y) = p(y)y = ay - by2

and soMR(y) = a - 2by < a - by = p(y) for y > 0.

Marginal RevenueE.g. if p(y) = a - by then R(y) = p(y)y = ay - by2

and soMR(y) = a - 2by < a - by = p(y) for y > 0.

p(y) = a - bya

ya/bMR(y) = a - 2by

a/2b

Marginal CostMarginal cost is the rate-of-change of totalcost as the output level y increases;

MC ydc y

dy( )

( ).

E.g. if c(y) = F + y +y2 then

MC y y( ) . 2

Marginal Cost

Fy

y

c(y) = F + y + y2

$

MC(y) = + 2y

$/output unit

Profit-Maximization; An ExampleAt the profit-maximizing output level y*,MR(y*) = MC(y*). So if p(y) = a - by andc(y) = F + y + y2 then

MR y a by y MC y( *) * * ( *) 2 2

Profit-Maximization; An ExampleAt the profit-maximizing output level y*,MR(y*) = MC(y*). So if p(y) = a - by and ifc(y) = F + y + y2 then

MR y a by y MC y( *) * * ( *) 2 2

and the profit-maximizing output level is

yab

*( )

2

Profit-Maximization; An ExampleAt the profit-maximizing output level y*,MR(y*) = MC(y*). So if p(y) = a - by and ifc(y) = F + y + y2 then

MR y a by y MC y( *) * * ( *) 2 2

and the profit-maximizing output level is

yab

*( )

2

causing the market price to be

p y a by a bab

( *) *( )

. 2

Profit-Maximization; An Example$/output unit

y

MC(y) = + 2y

p(y) = a - by

MR(y) = a - 2by

a


y

MC(y) = + 2y

p(y) = a - by

MR(y) = a - 2by

y

ab

*

( )

2

a


y

MC(y) = + 2y

p(y) = a - by

MR(y) = a - 2by

y

ab

*

( )

2

p y

a bab

( *)

( )

2

a

Monopolistic Pricing & Own-Price Elasticity of Demand

• Suppose that market demand becomes less sensitive to changes in price (i.e. the own-price elasticity of demand becomes less negative). Does the monopolist exploit this by causing the market price to rise?


MR yd

dyp y y p y y

dp ydy

p yy

p ydp y

dy

( ) ( ) ( )( )

( )( )

( ).

1


MR yd

dyp y y p y y

dp ydy

p yy

p ydp y

dy

( ) ( ) ( )( )

( )( )

( ).

1

Own-price elasticity of demand is

p yy

dydp y

( )( )


MR yd

dyp y y p y y

dp ydy

p yy

p ydp y

dy

( ) ( ) ( )( )

( )( )

( ).

1

Own-price elasticity of demand is

p yy

dydp y

( )( ) so MR y p y( ) ( ) .

11


MR y p y( ) ( ) .

11

Suppose the monopolist’s marginal cost ofproduction is constant, at $k/output unit.For a profit-maximum

MR y p y k( *) ( *)

1

1

which isp y

k( *) .

1 1


p yk

( *) .1 1

E.g. if = -3 then p(y*) = 3k/2, and if = -2 then p(y*) = 2k. So as rises towards -1 the monopolistalters its output level to make the marketprice of its product to rise.


Notice that, since MR y p y k( *) ( *) ,

1

1

p y( *) 11

0



1

1

p y( *) 11

0 11

0



1

1

p y( *) 11

0 11

0

That is,1

1



1

1

p y( *) 11

0 11

0

That is,1

1 1

.



1

1

p y( *) 11

0 11

0

That is,1

1 1

.

So a profit-maximizing monopolist alwaysselects an output level for which marketdemand is own-price elastic.

Markup Pricing

• Markup pricing: Output price is the marginal cost of production plus a “markup.”

• How big is a monopolist’s markup and how does it change with the own-price elasticity of demand?

Markup Pricingp y k p y

k k( *) ( *)1

1

11 1

is the monopolist’s price.

Markup Pricingp y k p y

k k( *) ( *)1

1

11 1

is the monopolist’s price. The markup is

p y kk

kk

( *) .

1 1

Markup Pricing

p y k p yk k

( *) ( *)11

11 1

is the monopolist’s price. The markup is

p y kk

kk

( *) .

1 1

E.g. if = -3 then the markup is k/2, and if = -2 then the markup is k. The markup rises as the own-price elasticity of demand rises towards -1.

A Profits Tax Levied on a Monopoly

• A profits tax levied at rate t reduces profit from (y*) to (1-t)(y*).

• Q: How is after-tax profit, (1-t)(y*), maximized?




• A: By maximizing before-tax profit, (y*).




• A: By maximizing before-tax profit, (y*).

• So a profits tax has no effect on the monopolist’s choices of output level, output price, or demands for inputs.

• I.e. the profits tax is a neutral tax.

Quantity Tax Levied on a Monopolist

• A quantity tax of $t/output unit raises the marginal cost of production by $t.

• So the tax reduces the profit-maximizing output level, causes the market price to rise, and input demands to fall.

• The quantity tax is distortionary.


$/output unit

y

MC(y)

p(y)

MR(y)

y*

p(y*)


$/output unit

y

MC(y)

p(y)

MR(y)

MC(y) + tt

y*

p(y*)


$/output unit

y

MC(y)

p(y)

MR(y)

MC(y) + tt

y*

p(y*)

yt

p(yt)


$/output unit

y

MC(y)

p(y)

MR(y)

MC(y) + tt

y*

p(y*)

yt

p(yt)

The quantity tax causes a drop in the output level, a rise in the output’s price and a decline in demand for inputs.


• Can a monopolist “pass” all of a $t quantity tax to the consumers?

• Suppose the marginal cost of production is constant at $k/output unit.

• With no tax, the monopolist’s price is

p yk

( *) .1


• The tax increases marginal cost to $(k+t)/output unit, changing the profit-maximizing price to

• The amount of the tax paid by buyers is

p yk tt( )( )

.

1

p y p yt( ) ( *).


p y p yk t k tt( ) ( *)( )

1 1 1

is the amount of the tax passed on tobuyers. E.g. if = -2, the amount ofthe tax passed on is 2t.Because < -1, ) > 1 and so themonopolist passes on to consumers morethan the tax!

The Inefficiency of Monopoly

• A market is Pareto efficient if it achieves the maximum possible total gains-to-trade.

• Otherwise a market is Pareto inefficient.

The Inefficiency of Monopoly$/output unit

y

MC(y)

p(y)

ye

p(ye)

The efficient output levelye satisfies p(y) = MC(y).


y

MC(y)

p(y)

ye

p(ye)


CS


y

MC(y)

p(y)

ye

p(ye)


CS

PS


y

MC(y)

p(y)

ye

p(ye)

The efficient output levelye satisfies p(y) = MC(y).Total gains-to-trade ismaximized.

CS

PS


y

MC(y)

p(y)

MR(y)

y*

p(y*)


y

MC(y)

p(y)

MR(y)

y*

p(y*)CS


y

MC(y)

p(y)

MR(y)

y*

p(y*)CS

PS


y

MC(y)

p(y)

MR(y)

y*

p(y*)CS

PS


y

MC(y)

p(y)

MR(y)

y*

p(y*)CS

PS


y

MC(y)

p(y)

MR(y)

y*

p(y*)CS

PS

MC(y*+1) < p(y*+1) so bothseller and buyer could gainif the (y*+1)th unit of outputwas produced. Hence the market is Pareto inefficient.


y

MC(y)

p(y)

MR(y)

y*

p(y*)

DWL

Deadweight loss measuresthe gains-to-trade notachieved by the market.

The Inefficiency of Monopoly

$/output unit

y

MC(y)

p(y)

MR(y)

y*

p(y*)

ye

p(ye) DWL

The monopolist produces less than the efficient quantity, making the market price exceed the efficient market price.

Natural Monopoly

• A natural monopoly arises when the firm’s technology has economies-of-scale large enough for it to supply the whole market at a lower average total production cost than is possible with more than one firm in the market.

Natural Monopoly

y

$/output unit

ATC(y)

MC(y)

p(y)

Natural Monopoly

y

$/output unit

ATC(y)

MC(y)

p(y)

y*MR(y)

p(y*)

Entry Deterrence by a Natural Monopoly

• A natural monopoly deters entry by threatening predatory pricing against an entrant.

• A predatory price is a low price set by the incumbent firm when an entrant appears, causing the entrant’s economic profits to be negative and inducing its exit.


• E.g. suppose an entrant initially captures one-quarter of the market, leaving the incumbent firm the other three-quarters.


y

$/output unit

ATC(y)

MC(y)

DI

DE

p(y), total demand = DI + DE


y

$/output unit

ATC(y)

MC(y)

DI

DE

pE

p(y*)

An entrant can undercut theincumbent’s price p(y*) but ...



y

$/output unit

ATC(y)

MC(y)


DI

DE

pE

pI

p(y*)

An entrant can undercut theincumbent’s price p(y*) but

the incumbent can then lower its price as far as pI, forcing the entrant to exit.

Inefficiency of a Natural Monopolist

• Like any profit-maximizing monopolist, the natural monopolist causes a deadweight loss.

y

$/output unit

ATC(y)

p(y)

y*MR(y)

p(y*)

MC(y)

Inefficiency of a Natural Monopoly

y

$/output unit

ATC(y)

MC(y)

p(y)

y*MR(y)

p(y*)

p(ye)

ye

Profit-max: MR(y) = MC(y) Efficiency: p = MC(y)


y

$/output unit

ATC(y)

MC(y)

p(y)

y*MR(y)

p(y*)

p(ye)

ye

Profit-max: MR(y) = MC(y) Efficiency: p = MC(y)

DWL


Regulating a Natural Monopoly

• Why not command that a natural monopoly produce the efficient amount of output?

• Then the deadweight loss will be zero, won’t it?

y

$/output unit

ATC(y)

MC(y)

p(y)

MR(y)

p(ye)

ye


At the efficient outputlevel ye, ATC(ye) > p(ye)

ATC(ye)

y

$/output unit

ATC(y)

MC(y)

p(y)

MR(y)

p(ye)

ye


At the efficient outputlevel ye, ATC(ye) > p(ye)so the firm makes aneconomic loss.

ATC(ye)Economic loss


• So a natural monopoly cannot be forced to use marginal cost pricing. Doing so makes the firm exit, destroying both the market and any gains-to-trade.

• Regulatory schemes can induce the natural monopolist to produce the efficient output level without exiting.

Documents

1 Economie Publique II February-May 2010 Prof. A. Estache Lecture 1 Overview