Management and Markets

1
Management and Markets

• Centralization, Decentralization, and Incentives
• (usual thanks to George Baker)

2
Course Content (Strategy and Org Econ.)

• Overview Introduction to Strategy
• Market Competition (sessions 2-6)
• Theory of the Firm (sessions 7-8)
• Internal Organization Incentives and
  Performance Management (sessions 9-11)
• Summary and Application to Innovation

3
Firms and MarketsTheory and Evidence

• Coase Firms exist where markets fail
• Williamson Markets fail because of hold-up
  institutions minimize hold-up (and other
  transactions costs)
• Grossman-Hart The costs and benefits of
  integration and hold-up

4
Grossman-Hart recap

• Hold-up ex post. The holder of the residual
  rights of control will have more power
• To whom should this power be allocated?
• Whoever has more power ex post will have stronger
  incentives to invest in the relationship
• GH bottom line
• Give the residual rights (ownership) to the party
  whose ex ante investment is more valuable

5
Is PRT a Theory of the Firm?

• Good for asset ownership, not so good for how
  residual rights are allocated in firms
• GH is really a theory of ownership
• Need a Theory of the Firm, to go along with a
  theory of ownership

6
How are firms organized?

• Centralization vs. decentralization
• Chandler, Strategy and Structure
• This affects many aspects of firms policies,
  practices, culture, etc.
• Two versions of the theory
• Jensen and Meckling
• Milgrom and Roberts
• Both start from the markets vs. firms question

7
Start with another classic Hayek

• What is the problem we wish to solve when we try
  to construct a rational economic order?
• The efficient allocation of known resources?
• Not!
• The use of knowledge in society

8
Planning

• The question is not whether an economy should be
  planned
• but rather who should do the planning
• Individuals and firms plan all of the time
• Centralized authorities or decentralized agents?

9
Knowledge

• Scientific vs. particular
• Specific to time and place
• Dispersed and fleeting
• Difficult to transmit
• Impossible to aggregate
• Arbitrageur-- Viewed with contempt/dishonest

10
Role of the market

• A decentralized price system makes efficient use
  of particular knowledge
• No central planner could do this
• Analog to organizations?

11
Jensen and Meckling 1992

• How are decision rights allocated in a market?
• How are they allocated in a firm?
• In a firm, they need to be allocated by top
  management
• Distinction between decision making and decision
  control

12
Considerations in the Allocation of DRs

• Specific knowledge
• Expensive to transmit
• Dispersed throughout the organization
• Necessary to good decision making
• Organizational objectives
• People are self-interested
• They cannot be counted on to always do the right
  thing

13
The optimal allocation of DRs

• Good decision making requires that the decision
  maker have both good information and the right
  objectives
• Move the information to those with the right
  incentives, or move the incentives to those with
  the specific knowledge
• Centralized vs. decentralized

14
Costs
Costs of imperfect incentives
Information costs
Degree of decentralization
15
Costs
Total costs
Costs of imperfect incentives
Information costs
Degree of decentralization
16
The optimal degree of centralization

• Balances the two types of costs
• At the optimum, both problems will exist
• The solution is second best optimal
• What determines the degree of centralization or
  decentralization?
• Type of decision
• Size of the firm
• Rate of environmental change
• Availability and efficiency of IT
• Nature of knowledge

17
Milgrom and Roberts on Centralization-Decentraliza
tion

• Again, start with the market
• Whats wrong with a purely decentralized system?
• Prices do not always work
• Impossibility of perfect Arrow-Debreu contingent
  contracts
• Why you need firms to capture economies of scale

18
MR Criteria for Efficient Coordination

• Can system identify efficient decision if perfect
  information is available to coordinator?
• How much communication is required to identify
  optimal decision?
• If information is imperfect, how badly does
  performance deteriorate?

19
General Resource Allocation v. Coordination
Problems (MR)

• Prices economize when nothing known prior to
  making allocation (general resource allocation)
• Coordination through centralized dictates
  economize when problem has design attributes
  (parts of decision must fit together and big
  costs to failures of fit--brittleness) and
  coordinator has information about these design
  attributes
• Strongly complementary activities (e.g.
  sychronous problems the crew example)
• Assignment problems (need just oneambulance
  example)

20
Costs of decentralization

• How can decentralized decision-makers be
  coordinated?
• Can we simply set up an incentive system and let
  the decentralized agents do what is in their own
  best interest?

21
Costs of centralization

• A centralized decision maker (with authority)
  lacks all of the information necessary for good
  decisions
• Needs to get information from subordinates
• But subordinates have an interest in the
  decisions that she will make

22
Influence costs (MR)

• Subordinates will try to manipulate the decision
  maker to induce decisions that are in his best
  interest
• The biases may lead to bad decisions
• There are direct costs of the lobbying
• Rent seeking
• Fighting over the distribution of the pie, rather
  than trying to make the pie bigger

23
The Arrow Selling Ideas Problem

• How is decision information transmitted when
  transmitter has incentive to misrepresent
  information and receiver has incentive to
  expropriate information?
• Problem both across organizations and within
  organizations
• Moving the decision to holder of information?
• Inefficient organizational use of information?
• Anton-Yao (94) information problem with no
  contracting protections
• One solution turn adverse selection
  (information) problem to moral hazard
  (expropriation) problem and partially solve via
  competition (blackmail)
• AY95 applies this to ownership model regarding
  invention

24
The centralization-decentralization trade off

• Where prices do a particularly bad job of
  coordinating, it may be worth bearing the
  influence/agency costs and centralizing
• Where information about production is
  particularly impacted, organizations will
  decentralize
• Where objectives are particularly mis-aligned,
  organizations will centralize

25
Incentive alignment and the agency problem

• Any organization needs three systems (JM)
• System to allocate decision rights
• Performance measurement system
• Reward and punishment system
• These are conceptually distinct systems
• But efficiency dictates certain relationships
  between and among them

26
Next class Incentive Systems

• What are the problems with designing systems to
  control decentralized decision rights?
• Performance measurement
• The folly of rewarding for A while hoping for B
• Reward systems
• Risk
• Motivation

27
Questions

• Relating theories
• In the property-rights theory of ownership
  location of information is not prominently
  discussed. JM make location of information a
  focal point of their theory of the firm. How do
  these two approaches relate to one another?
• Alchian and Demsetz seem to suggest that there is
  little (except team production) that
  differentiates activity in a firm from that in
  the market. How does this relate to JMs view
  of alienability?
• Given strategies of various economic actors (e.g.
  suppliers), how well does the price system truly
  reflect social scarcity of resources? How do
  failures of alienability relate to market
  failures?
• How doest this economic organization logic square
  with theories of scientific management (e.g.
  Taylor or Gilbreth)?

28
Questions

• Applications
• How has information era changed the nature of
  types of knowledge and markets for knowledge?
• Are decision rights captured by those who value
  them most highly? How well does the
  MA/strategic alliance market work for knowledge
  acquisition?
• Management and Strategy
• How does uncertainty and bounded rationality
  about the future consequences of ones decisions
  undermine, if at all, the benefits of
  alienability of decision rights?
• How does corporate culture and organizational
  politics work in the theories of JM and MR?
• How should we think of JM and MR in the context
  of creative versus production-type organizations?
• Arent there market fixes (e.g. confidentiality
  agreements) to problems such as lack of property
  rights or other informational problems both
  within and between organizations?