Competition Policy 2

1
Competition Policy 2

• Regulation of market power

2
Goals

• To understand
• How regulatory regimes impact on incentives?
• The economic benchmarks for price regulation
• Practical issues associated with the
  implementation of price regulation

3
Foundational Issue I

• Why regulate?

4
When is regulation needed?

• Externalities
• Lack of competition
• Natural monopoly
• when it less costly to have all output supplied
  by one firm than by more than one firm
• examples in electricity, transport,
  telecommunications
• the role of competition
• is gas distribution a natural monopoly?
• does it require regulation?

5
Types of regulation

• Consumer pricing
• Regulator sets tariffs to consumers
• Access pricing
• Regulator sets prices for access to essential
  infrastructure
• Assesses need on uneconomic to duplicate
• Intervenes if negotiations fail

6
Foundational Issue II

• How does regulation impact on incentives?

7
An Exercise

• Suppose you are the regulator of a rail network
• At present, users can utilise the network paying
  a price of 200 per car-km along the network
• You receive verifiable information that each
  additional car-km costs the firm 150, given that
  the rail network is already constructed
• You have legal power to set a new price at any
  level that you choose
• What price will you choose?

8
Trade-Offs

• What are the benefits of pricing low (close to
  150)?
• What are the benefits of not pricing low?

9
Performance-Based Regulation

• A procurement example suppose you want to
  purchase cleaning services for your office.
• Your choices ...
• On-going cost-plus contract
• Medium-term fixed price contract

10
Cost-Plus

• Benefits
• Increased flexibility to changed circumstances
• No rents accruing to supplier
• Costs
• High risk if dont know costs (adverse selection)
• Even if can observe costs, supplier has no
  incentive to invest in cost reduction (moral
  hazard)
• Involves low-powered incentives for every 1 in
  cost reduction achieved, the supplier receives 1
  less in profits

11
Fixed-Price

• Costs
• Are you getting value for money?
• Supplier bears risk
• Benefits
• Can get a lower price initially that is, if
  costs are currently 10 but you anticipate that
  next year (if investments are made), they could
  fall to 8, you could get a price of (say) 9
  over the two years.
• Involves high-powered incentives for every 1 in
  cost reduction achieved, the supplier gets 1
  extra in profits

12
Incentive Trade-Offs

• Do you get value for money?
• Can you commit not to renegotiate price terms if
  costs fall at a rapid rate
• Increased risk placed on the supplier
• What about non-price issues (e.g., quality and
  reliability of service)?

13
Basic Trade-Off I

• A high-powered incentive scheme involves leaving
  rents with the agent
• How do you avoid temptation?

14
Procurement again

• Can the customer get out of the fixed price
  agreement?
• Raise quality concerns
• Withhold payment
• Contracts may not be complete, leading to the
  potential for ex post opportunism, hold-up or the
  ratchet effect

15
Overcoming the Ratchet Effect

• Lock-in customer to the supplier
• Co-specialised investments
• Example Apples Mac plant in 1984
• Stipulated damages
• Arbitration
• Relational contract
• If relationship is long-term, then a far-sighted
  customer will refrain from renegotiation
• Customers reputation as a good client

16
Regulatory Commitment

• Is there potential for regulatory opportunism?
• Regulatory discretion
• Interpretation of legislation
• Political pressure
• Government can change legislation
• Impose other regulations
• Short-term government outlook

17
Incentives and Insurance

• Basic message dangerous to insure suppliers
  against uncontrollables
• Examples
• plant manager who has decision rights over
  production process but no control over the
  price of raw materials. Should this price be
  removed from performance measure? If it is
  removed then manager has reduced incentive to
  alter production process in face of changes in
  price.
• Truck fleet manager whose performance measure
  does not include fuel cost has no incentive to
  economise on fuel when oil prices double
• Be careful if design performance measurement that
  gives suppliers blanket insurance policies. Can
  cause large distortions in incentives.

18
Basic Trade-Off II
Pure Incentive Regulation
Pure Cost-Based Regulation
0
100
Extent of Earnings Insurance
Source adapted from Ergas-Small (2001)
19
Quality Issues

• Is it a good idea to ...
• Evaluate airline pilots on number of on-time
  flights?
• Evaluate a typist on number of typed words?
• Evaluate government employment agency workers on
  proportion of job applicants who found jobs?
• Reward SA Rugby Union players on number of
  tackles of Lomuh?
• Evaluate teachers evaluated on test scores?
• Reward academics rewarded for number of papers
  published?

20
Basic Trade-Off III

• Care must be taken in specifying performance
  measures in high-powered incentive schemes so as
  to avoid omission of critical performance criteria

21
Summary

• All regulatory regimes set incentives
• High-powered incentives can result in mutually
  beneficial outcomes for regulated firms and
  consumers
• Issues
• Can the regulator commit to the incentive scheme?
  Strong (expropriation) versus weak (firm
  insurance)
• Can the regulatory regime specify the key
  performance criteria?

22
Pricing principles

• Upon what basis should regulators set prices?

23
Information used

• Cost-based
• Allow cost recovery and rate of return
• Provide good signals to consumers
• But value loss if only linear prices can be
  used
• Demand-information
• Take into account customer elasticities
• Ramsey pricing recover more costs from inelastic
  customers
• Example peak-load pricing

24
Practical Aspects of Regulation

• How do you implement sound price regulation?

25
An Historical Approach

• Early experience in UK - regulation by parliament
• Regulatory commissions
• Smyth v Ames and fair return
• US rate-of-return regulation
• Modern developments
• UK privatisations and price caps
• deregulation in the US
• NCP and regulation in Australia

26
US-Style Rate of Return Regulation

• P (rK OM D)/QEST
• r allowed return on capital K
• D depreciation
• QEST estimated sales at regulated price
• Prices fixed between reviews
• May involve unders and overs accounts - low
  firm risk

27
RoR Problems

• Purely cost based
• low incentives to lower cost
• intrusive data collection by regulator
• Incentive to expand rate base
• prudency reviews and used and useful asset test
• Multiple-products, competition and cost-shifting
  - the ATT example
• Regulatory capture

28
Price caps - origins

• Builds on traditional price controls by allowing
  automatic price adjustment
• Recommended for BT by Littlechild in 1983 -
  viewed as transitional
• Applied to privatised BT in 1984
• Is now commonly applied in some form world-wide

29
Price caps Theory (one product)

• P1 (1 CPI - X).P0
• Fixed period between reviews (eg. 5 years)
• Strong incentives to reduce costs
• CPI - X shares expected gains
• CPI captures economy wide productivity gains and
  is a proxy for input price rises
• X reflects expected idiosyncratic productivity
  gains

30
Price caps Theory (multiple products)

• Basket of prices
• SP1Q0 (1 CPI - X). SP0Q0
• i.e. use last years quantities as weights
• rebalancing can only make consumers better off in
  aggregate
• over time can lead to Ramsey prices
• can include non-linear tariffs

31
Price caps problems

• Need quality controls
• Re-balancing and competition BT and Mercury
• Use of cost pass-through British Gas
• What if estimate X incorrectly
• regulatory commitment Littlechild and
  electricity
• Resetting the price cap

32
Price caps practice

• Resetting the price cap has generally involved a
  US style rate review.
• Cost savings are clawed back reducing
  incentives
• Reviews have involved intrusive, firm-specific
  data collection
• In Australia, X is not related to expected
  productivity
• Is it just RoR with a regulatory lag?

33
US Changes since the 1980s

• Moving away from traditional RoR regulation
• competitive entry and asset stranding
• price caps
• TSLRIC and multiple product firms
• individual product price is set at the long run
  incremental cost plus an allocation of common
  costs
• e.g. 1996 Telecommunications Act

34
Better regulation using benchmarks

• Example suppose five similar firms operate in
  different regions. At review, each firm reports
  average cost to regulator. The regulator then
  sets each firms price on the basis of other
  firms reports.
• Each firm keeps all of own cost savings - strong
  incentives
• Little incentive to distort report
• Induced competition improves allocative
  efficiency

35
Issues in benchmarking

• Collusion in reporting
• Firm idiosyncrasies - can use sophisticated
  econometric techniques to allow for imperfect
  correlation
• Using comparative analysis to set X under a price
  cap - for example TFP analysis

36
Australian Experience
37
Australian Experience

• Variety of industry specific regimes
• Almost all cost based
• Emphasis on establishing a capital base for
  newly privatised firms (eg., DORC, ODV, DAC)
• Concentration on RoR formulae
• Different approaches across jurisdictions (eg.
  ORG as highly intrusive, IPART ad hoc approach)
• Little to no use of non-linear prices
• Little to no use of Ramsey prices

38
ORG on Rail

• Simple uniform pricing
• Solely marginal cost based
• No inclusion of investment (leasing) costs
• Much of the network was constructed decades ago
  and can be regarded as a sunk cost for which no
  capital return is required.... For the same
  reason no amount is allocated for rent or other
  payments made by the Access Provider under its
  lease.
• High regulatory risk resulting in poor investment
  incentives

39
ORG on Electricity

• Rate-of-return approach under a price cap
• Ratchet effect institutionalised on 5 year basis
• Highly intrusive firm specific cost measurement -
  failure to use benchmarks properly despite
  rhetoric
• No attempt to implement efficient X in CPI - X
  approach
• No attempt to set Ramsey prices
• In contrast, ACCC has recommended non-linear
  pricing for electricity transmission (albeit
  based on previous use)

40
ACCC on Telecommunications

• Access prices on basis of TSLRIC
• Complex high-level modeling of network
• Different approaches to different prices (eg.
  basic access v local call resale)
• Long time delays

41
Summary

• Practice of price regulation is far from
  economists theoretical ideal
• Arguing for performance-based regulation is a
  hard sell
• Requires commitment in the face of monopoly rents
• Requires commitment in the face of poor firm
  performance
• But there has been improvement
• Rarely see purely cost based regulation (cf ORG
  on rail and ACCC on telecommunications)
• More discussion of non-linear pricing alternatives