4th Lecture, STV4346B: States and markets II

1
4th Lecture, STV4346B States and markets II

• Carl Henrik Knutsen, Department of Political
  Science, UiO
• 20/11-2008

2
Przeworski Ch.5. Politics, coordination and
degree of conflict

• Different structures for different political
  problems
• 1) Harmonious interests, pure coordination,
  reducing uncertainty and finding best solution
• 2) Mixed sum games. For example Prisoners
  Dilemma-structured situations. Some mutual gains
  exist, and coordination generates better outcome,
  but also distributional conflicts and incentives
  to deviate from cooperation
• 3) Pure distributional conflicts. Some win and
  some loose
• Degree of enforcement and coercion necessary to
  implement solution increases from 1-3.
• Does 1, 2 or 3 dominate in politics? Empirical
  question

3
Objectives of political actors

• No clear cut goal for political actors, like
  utility from consumption in consumer theory or
  profits in producer theory.
• The more sociology we enter into our analysis,
  the better the economic apparatus works..Optimal
  action is a function of constraints and
  objectives a a(C, U), but we only observe C.
  We have to make assumptions on U? Contextualized
  assumptions necessary.

4
Some potential objectives

• Personal political power
• Personal wealth
• Personal prestige (ego rents)
• Social welfare
• Interest group welfare
• Ideological vision or specific policies
• Empirical complexity Actor has more than one
  objective, and different actors put different
  weights on different objectives

5
Types of states a comparison

• Neo-patrimonial vs bureaucratic states (think
  Communist)
• If social welfare optimizer U U(Y).
• However, political leaders might have preferences
  over the amount of resources spent on public
  capital, G? UU(Y,G)? Will no longer allocate
  public capital so that maximize Y.
• Analysis, self-interested political leaders
• Neo-patrimonial leaders will underinvest in G,
  because money taken from their personal
  coffers. U(G)lt0
• Bureaucratic leaders can not privately steal
  public funds, but can use public capital
  personally without property entitlement? Build
  large public houses for personal use, use public
  jets etc? U(G)gt0? Will overinvest in public
  capital

6
A general, multiple-layered principal agent model
of politics
7
Implications

• Generically impossible to ensure that the will
  of the people is embedded in final regulation.
• The reasons are the inherent informational
  problems in the principal-agent relations
  combined with personal objectives of the
  different agents at different stages.
• Different adverse selection and moral hazard
  problems
• State failures and second-best solutions How
  to mitigate problems

8
Ch 6 Regulation

• Dont worry too much about the calculations..Read
  through chapter quickly.
• Def. regulation All actions by the state
  designed to cause some specific actions on the
  part of individuals or some specific allocation
  of goods or services
• Regulation under uncertainty
• More specifically regulates under asymmetric
  information. Economic actors often have more info
  (about own type and effort) than bureaucrats? use
  this info to generate rents.
• Groups will seek to make politicians/bureaucrats
  regulate in their favor ? endogenous regulation

9
Regulation contd

• What if regulating agency colludes with the
  object it is supposed to regulate, to capture
  mutually beneficial gains at the cost of wider
  society? (regulatory capture)
• Example Monopoly. Socially efficient regulation
  set regulated price so that equals marginal cost.
• If asymmetric info Monopoly lies about marginal
  cost-? earns extra profit..but sometimes risk
  that the agency finds out if uses resources to
  monitor
• If regulatory capture agency and monopoly
  collude and lie to politicians about cost, and
  split profit
• Remedies to such problems? External monitoring,
  competition, comparative benchmarks, fines if
  detected, installation of bureaucratic norms
  against collusion, give rents to firms so that
  they reveal correct info (costly, but perhaps
  better than initial situation)

10
Money and politics

• 1)Does money buy policies?
• Do politicians sell regulation to special
  interest groups?
• 2)Does money buy votes?
• Are voters sensitive to campaign spending, (plus
  patron-client relations, a perhaps more serious
  phenomenon)
• Campaign contributions can reflect both of the
  above factors 1) make deals with politicians, 2)
  give money to preferred politicians for use to
  influence impressionable voters (politics and
  fluid preferences, lack of information, rational
  ignorance, other cognitive and emotional factors)

11
Ch 7. Oversight

• Politics, universalism and partisanship Parties
  elected by segments of population, but intended
  to create rules that apply for all.
• Partisan control over bureaucracy comparative
  differences
• Crucial question How to avoid personal use of
  public power/office?
• Balancing features PR and coalition government,
  checks and balances, contramajoritarian
  institutions (constitutions, rights, judiciary)

12
Delegation

• Complexity of politics, incomplete contracts and
  asymmetric info? Can not specify behavior
  precisely.
• What are the goals, how are tasks done, lack of
  information feedback? Rely on bureaucratic norms
  and rules rather than discretionary orders. More
  on this in Fukuyama.
• Information advantages If bureaucrats behave,
  they will act more efficiently than politicians
  higher up in the command chain. Local knowledge
  and specialization
• What if bureaucrats do not behave? Fire alarms,
  citizens and NGOs. Threats of substantial
  punishment if cheating detected?

13
Checks and balances

• Potential damaging actions far worse than just
  shirking on effort..Corruption, abuse of
  authority etc..
• Institutional checks at least one other agent
  that can block actions.
• Benefits, but also drawbacks (rigidity)
• Same goes for rule following behavior vs
  discretionary behavior Larger risks with
  discretionary behavior, but rules can cause
  rigidity. Trade-off.
• Rule of law read Przeworskis discussion. The
  liberal idea of constraining individual behavior
  in politics.
• Bureaucracies or committees with agenda setting
  power? Given Arrow/McKelvey problems, this has de
  facto implications for end result.
• Politicians with preferences close to the
  bureaucrats might choose rationally to delegate.
  This mechanism is even stronger if take into
  account that other parties might occupy power in
  next period.

14
Ch 8 Representation

• Why may (democratic) governments act in peoples
  best interest?
• Self-selection of the publically spirited to
  politics
• People have at least some level of info about
  politicians vote on the good guys
• Retrospective voting and the threat of being
  thrown out of office
• Checks and balances constrain negative behavior

15
Prospective vs retrospective voting

• The benefits of retrospective voting Politicians
  have an incentive to generate better outcomes for
  society as a whole (accountability issues)
• The role of swing voters
• The role of sufficient information and
  transparency (complexity of causal mechanisms as
  a problem for eliciting info)
• The drawbacks of lameduck sessions
• When voters act prospectively and vote on the
  politician they believe will do best in the next
  period, they reduce incentives for incumbents to
  behave.
• Links between retrospective and prospective
  voting Bayesian updating You learn about the
  incumbent from observing policies ? retrospective
  and prospective voting might coincide
• The problem of multiple policies and few votes
  My party did well on reducing unemployment but
  bad on the environment..What shall I do?
• The role of the opposition as an information
  provider on the incumbents policies, but
  credibility issues when reporting..

16
Fukuyama ch2.

• Read through rapidly!! Central points are on
  these slides, or are summed up in other parts of
  the curriculum
• General claim No single best practices when it
  comes to public administration! Contextual needs
  and interaction effects..
• But not relativism Range of good practices and
  range of bad practices. However, no universal
  ordering.
• Conundrum of organizational theory Efficiency
  requires delegated discretion, but open up for
  control and supervision problems
• Efficiency and the role of local knowledge
  Friedrich von Hayeks argument against planned
  economies and centralization.
• The firm and institutional economics, Williamson.
  Factors that determine the relative efficiency of
  trading in markets vs within firms
• Bounded rationality, complexity of transact. and
  incomplete contracts
• Power and efficiency (specialized production and
  few actors),
• Uncertainty? Organize activities within firm to
  reduce transaction costs

17
Organizations and problems of principal-agent
theory and first best solutions

• Unclear goals of organizations (what should or do
  hospitals or schools maximize?)
• Lack of specificity of activity? costly to
  monitor or control agents? reliance on informal
  norms as a better approach
• Optimal degree of delegation is very context
  dependent (sector, time period, type of actor
  etc)
• Political decentralization and local knowledge.
  The subsidiarity principle.
• But coordination problems and the question of
  local administrative capacity.

18
Informal norms

• Endogenous preferences the role of leadership,
  tradition, team spirit.. Preference-shaping as
  powerful tool.
• But decentralized shaming and praise, as well as
  career ladders, within the agency as powerful
  incentives for individuals.
• What is appropriate behavior? (logic of
  appropriateness rather than maximization)
• Low degree of specificity, low degree of clear
  informational feedback, lack of comparative
  yardsticks? rules of thumb
• Formal incentive structures will work best in
  cases where low transaction volume and high
  degree of specifity (central banking vs primary
  education)

19
Evans ch. 3

• Bureaucratic structure and the benefits of
  industrial policy. The effect of industrial
  policy depends on the structure of the
  bureaucracy and state-society relations
• Read the case-descriptions in Evans yourselves,
  and at least a couple of them thoroughly. We will
  get back to some of these countries in the last
  part of the course
• Bureaucracies
• Recruitment of the best and brightest
• The internal coherence in bureaucratic agencies
• The importance of a Weberian structure, the role
  of norms. Self-interest and maximization without
  constraints as a recipe for failure.

20
Evans

• Autonomy and embeddedness
• State autonomy as ability to shape and implement
  own goals, and ability to solve collective
  problems.
• But reliance on informational feedback and the
  need to be embedded in society for policy to be
  efficient. Information and cooperation versus
  capture by social interests.
• Predatory vs developmental regimes

21
Webers ideal model

• 1) Jurisdictional areas are clearly specified,
  activities are distributed as official duties
• 2) Hierarchical organization Subordinates follow
  orders from superiors, but have appeal rights.
• 3) Relatively stable rules govern decisions and
  actions.
• 4) Personal property is separated from office
  property.
• 5) Officials are selected on basis of
  qualifications, and are appointed (not elected).
  Meritocracy!
• 6) Officials are compensated by salary (should
  not be too low to reduce probability of corrupt
  behavior).
• 7) Employment is a (life-long) career. Officials
  are protected from arbitrary dismissal. The
  motivational factor of the career ladder

22
Evans and Rauch

• What is the effect from a Weberian
  bureaucracy-structure on economic growth?
• Construct data from expert surveys, but fixed,
  specific questions
• (Problems of endogenous scoring on open
  questions Since Singapore has produced a high
  growth rate, its bureaucracy must be good lets
  give it a high score!!)
• 35 developing countries from 1970-1990
• Inertia in bureaucratic structures ? Do not use
  the time dimension
• Use two particular dimensions from the Weberian
  model Meritocratic recruitment and long term
  careers/career-ladders. Are these two dimensions
  correlated with other dimensions of Webers
  model? Probably.

23
ER contd

• The focus on institutions and governance as
  determinants of development in latter years
• But a need for specification of theory and better
  data
• Weber and the bureaucracy as a tool for growth
• selection of good policy/regulation
• stability
• exit patrimonialism and personal rule
• Case studies that suggest the relevance of the
  above hypothesis Johnson on Japan, Amsden on
  South Korea, Wade on Taiwan

24
More specific links between the two dimensions
and growth

• Meritocracy (M) and quality of bureaucrats ?
  better policy
• Career stability (CS) and regulatory stability?
  risk reduction for private actors
• CS and competence ? better policy
• MCS ?esprit de corps ? less corruption
• CS? longer time horizons and incentives to behave
• Coherent, competent bureaucracies and
  coordination abilities plus problem solving
  abilities

25
Does it really matter empirically? Yes!
26
Regression results