Is Economic Regulation Possible

1
Is Economic Regulation Possible?

• Nick Wills-JohnsonCRAE PATREC
• PATREC Research Forum
• 4 September 2007

2
Presentation in a Nutshell

• Nobody here working for government in a role
  which decides resource allocations serves the
  public good, unless their world involves no
  Knightian uncertainty.
• A Parliament cannot, in general, design a
  resource allocation mechanism to be implemented
  by objective agents.
• Instead, you all impose your own will on society,
  or make choices which cannot be justified by
  reference to benefits to members of society in
  any specific way.

3
Outline

• Deciding resource allocations.
• Economic regulation objectivity.
• Arrows Impossibility Theorem.
• Arrow in economic regulation.
• Consequences of impossible regulation.
• Conclusions and policy recommendations

4
Deciding Resource Allocation

• Market
• Price and value decides allocation.
• No good for market failure cases.
• Political
• Negotiation or voting with no fixed rules.
• Bureaucratic
• Parliament creates a resource allocation
  function and bureaucrats administer it.
• Public sector regulation vs management.
• Convention something well return to later

5
Planned Actual Delegation
What you want
What you get
6
Economic Regulation in Operation

• Minimise monopoly rents
• Best practice model
• LRMC
• Process
• Proposed AA
• Structure of Building Blocks
• Consultation
• Draft Final Decision

7
Is it Objective?

• How objectively can you forecast?
• 1,3,5,7,..
• What comes next?
• 9
• 10
• 11
• 12

8
Is it Objective?

• Y1,3,5,7,9
• Y1,3,5,7,10
• Y1,3,5,7,11
• Y1,3,5,7,12
• Y1,3,5,7, 4532187
• Forecast lt assumptions lt preferences
• Regulation is social choice

9
Arrows Impossibility Theorem

• Tries to map from individual to social prefs.
• Five conditions
• Universality, monotonicity, IIA, non-imposition
  and non-dictatorship.
• Cant have all five at once (impossibility).
• End up with either dictatorship of a social
  preference which is unrelated to individual
  preferences.

10
Criticising Arrow

• Huge literature, two streams
• The fundamental underpinnings are incorrect
• Transferring notions of individual choice to a
  group is a fallacy of composition.
• We can relax some of the conditions
• Plott (1976) shows all the pathways explored.
• Impossibility stays pretty robust.
• If utility is cardinal and comparable, there is
  no impossibility.

11
Arrow in Regulation

• Three party game (owner, customer regulator)
• Parties form preference schedules
• Describe these in submissions to regulator
  DD/FD
• Utility described by information of two types
• Objective information.
• Subjective information.
• Revealed information builds forecasts.
• Means forecasts show preference schedules
• Type of preference schedule determines regulatory
  possibility
• If ordering of preference schedule det only by
  objective info gtOK, resource allocation function
  works
• If ordering of preference schedule det by
  subjective info gt Arrovian impossibility.

12
Consequences of Impossibility

• Arrovian dictatorship
• Regulator most likely dictator
• Industry unattractive to entry.
• Resistance to regulation from all stakeholders
• Unreflective social preferences
• Likely negotiated solution
• No PMC unless regulator lies
• Both of these outcomes bad.
• Hard to predict outcomes
• Entry less likely.

13
Regulatory Evolution

• Allow for strategic revelation of information
• Regulator as dictator
• Catch 22 more objectivity potential
  regulatory creep, less means more subjective
  regulation.
• Can a regulator maintain a light hand?
• Unreflective social preferences
• Attempts at regulatory capture
• Similar catch 22 of infrastructure socialisation.
• Potential cyclicality of regulation
• US, UK history (also for regulatory creep)

14
Policy Issues

• Arrow says four ways to allocate resources
• Market, vote, convention or dictatorship.
• Regulation is an attempt at a specific
  convention, which ends either in dictatorship by
  professionals or in a different convention
  decided in the industry/sector.
• Dictatorship might be OK but it is based on the
  regulators subjective beliefs, not objective
  skills.
• A different convention might be OK, but it could
  be subject to regulatory capture.

15
Policy Solutions?

• Use a market
• Wont always work (natural monopoly) and
  designing one faces same problems as regulation.
• But Sonnenschein axiomatic market benchmarks.
• Use a vote
• Complexities of plebiscites politicisation of
  parliaments.
• Use a different convention
• Stop listening to economists, read the tragedy of
  the commons literature.

16
Further Thoughts

• Comments and further thoughts
• n.wills-johnson_at_curtin.edu.au.
• More papers and other work
• http//ssrn.com/author106139