Incentives and Firm Boundaries

1
Incentives and Firm Boundaries

• John Roberts
• Stanford GSB
• Steyr
• July 2004

2
Why Firms?

• Coordinate and motivate
• But markets do this well!
• Market and contracting failures
• The firm is an alternative to the market that may
  perform better at coordinating and, especially,
  motivating in the presence of contracting
  difficulties.

3
Strategic Management 101, Lecture 1Determinants
of Performance
ENVIRONMENT
PERFORMANCE
STRATEGY
ORGANIZATION
4
Performance and Fit
ENVIRONMENT
PERFORMANCE
ORGANIZATION
STRATEGY
5
Environment

• Customers
• Competitors (actual and potential)
• Suppliers
• Complements and substitutes
• Technology
• Social, legal, regulatory, environmental
  government and NGOs

6
Strategy

• How will we win? How will we get control of our
  destiny?
• Business strategy
• How will we create value in this business and get
  to keep some of it?
• Corporate strategy
• What business portfolio? How will we create value
  by having these businesses under common ownership
  and our direction?

7
Organization

• The means through which strategy is realized and
  executed
• Structure follows strategy.--- Alfred Chandler
• Sets the context for strategy
• Organizational inertia means organization and
  administrative heritage constrain strategy
  choices
• Capabilities are embedded in the organization
• Organization shapes behavior and choice
• Our organization is our strategy. --- John
  Browne

8
Elements of Organization PARC

• People
• Architecture
• Routines, Processes and Procedures
• Culture

9
PARC People

• Their intelligence, skills and abilities
• Their motivations and fears
• Their attitudes to life, work and risk
• Their personal and professional interests

10
PARC Architecture

• Formal structure --- Whats on the org chart
• Grouping of tasks into jobs, jobs into units,
  units into departments/divisions,
• Hierarchic authority/reporting structure
• Allocation of decision rights/responsibility
• Boundaries of the firm
• Financing, ownership and governance
• Informal networks

11
PARC Routines

• Formal and informal managerial procedures,
  processes, routines that determine
• How work is done
• How information is gathered and disseminated
• How resources are allocated
• How performance is measured and rewarded
• How decisions are made

12
PARC Culture

• Shared values, beliefs, mental models, norms,
  language
• Why are we in business? What really matters?
• How do we act towards one another and outsiders
  in different situations? What can we expect from
  one another?
• How do we interpret observations and actions?
• What do we mean by various words?

13
PARC and Authority

• Ownership of assets (including contractual ones)
  gives power to design and enforce the rules of
  the game (Holmstrom 99, Rajan and Zingales)
• CEO has unique authority to select strategy,
  design PARC, take decisions (even if nominally
  delegated) (Milgrom-Roberts)

14
Two Classes of (Static) Theories of the Nature of
the Firm

• Create ex ante incentives
• GHM Property Rights
• HM Incentives in Multitasking
• Utilize ex post governance
• Williamson KCA Rent-seeking
• Hold-up or ex post haggling
• MR influence costs
• Simon Adaptation

15
PARC and the Boundaries of the Firm

• Strategic management focuses on what the firm
  does itself --- activities --- as defining the
  boundaries
• Organizational economics has tended to focus on
  what the firm owns --- assets --- as defining the
  boundaries.

16
PARC and Management

• The evidence from the field is that all the
  elements of PARC are used to provide motivation
  in firms, in part because there are so many
  complex behaviors to motivate
• Further, the features interact and so need to be
  considered holistically, as a system
• Good managers do this

17
PARC and Economics

• Most of the economics of organization has tended
  to isolate one aspect of the overall motivation
  problem and then examined how one aspect of
  organization can deal with it.
• E.g. Grossman-Hart-Moore, standard PA theory
• No real firms or cant tell if firm or market

18
Simplest GHM

• 2 parties, U and D, HC investments x and y
  respectively, costs z2/2. Non-contractable.
• One asset
• V(x,y)kUxkDy if cooperate, vU(x, j) and vD(y,
  j) if do not, where j indicates who has the
  asset vi(z,i)z, vi(z,j)0.
• Ex post bargaining, assumed efficient

19
Simplest GHM

• Bargaining gives payoffs that depend on asset
  ownership (and investments)
• If U owns (non-integration)
• vU(x,U)V(x,y) vU(x, U) vD(y,U)/2
• V(x,y) vU(x, U)/2 (kUxkDy x)/2.
• vD(y,U)V(x,y) vU(x, U) vD(y,U)/2
• V(x,y)/2 (kUxkDy)/2.
• So x(kU1)/2, ykD/2

20
Simplest GHM

• Similarly, if D owns (integration)
• xkU/2, y(kD1)/2
• Which is better depends on where most valuable to
  have more investment kU vs kD.
• Ownership here gives incentives for investment
  and so should be allocated to maximize the
  incentives.
• Cost of good incentives for one is bad ones for
  other party.

21
GHM

• Theory of asset ownership by individuals
• Not a satisfactory theory of the firm Principal
  conclusion is counter factual
• Firms, not individuals within them, own assets
• Is the relation between U and D under D-ownership
  that of employer-employee? Does this in any way
  resemble real firms?

22
Simplest Moral Hazard

• Contract with certain properties
• But looks the same whether inside or across
  boundaries
• So again not a theory of the firm.

23
Linear-Exponential-Normal Agency Model

• Linear contracts (w??x), exponential utility of
  income less cost C(e) of effort e (risk aversion
  coefficient r), x e normal noise (variance v)
• Contract satisfies

24
Multitasking

• A richer model arises from recognizing the
  multiplicity of behaviors that might be induced,
  multiple instruments
• Second best theory from public economics relevant
• Holmstrom-Milgrom 91, 94 Holmstrom 99

25
Inside or Outside Sales Force?

• Anderson, Anderson and Schmittlein 84
• Two tasks sell, gather info. First is easy to
  measure, second is hard. v1ltltv2, so if separate,
  ?1gtgt ?2.
• If all that matters is cost of effort, not its
  allocation, then if multitasking must have ?1
  ?2 ?M. Then ?M will be small (perhaps even less
  than for task 2 in isolation). Little effort.

26
Inside or Outside Sales Force?

• Now suppose a third action which exogenously
  offers a high incentive intensity (sell others
  product)
• If outside agent, cannot easily prevent taking
  this option, so must offer strong incentives to
  attract attention. Means no info gathering.
• If inside, can offer weaker incentives but ban
  doing third option.
• If info gathering important, use inside.

27
Employee or Contractor?

• Ownership of tools
• Effort on production (easy measure), maintaining
  tools (imprecise measure)
• If Worker owns tools, faces strong incentives for
  maintenance, must offer strong incentives for
  output.
• If Principal owns tools, can offer only weak
  incentives for maintenance, so must offer weak
  for output.

28
Ownership and Incentives

• Agent 1 uses single asset to produce y1 R(e1)
  and y2e2. 1 can sell y1 for price p1, but needs
  2 to sell y2 at p2. 2 has no investment decision.
  
• No contracts on ys, bargain ex post.

29
Ownership and Incentives

• If 1 owns, can get (1/2) p2y2 by threatening to
  withhold y2 plus p1R(e1) because does not need 2
  for this. With quadratic costs, sets
  e1e2p1R(e1)(1/2)p2. e1 above first best, e2
  below
• If 2 owns, then 1 gets (1/2)p1R(e1)p2y2, so
  e1e2(1/2)p1R(e1)(1/2)p2. e1 at FB, e2 closer
  to FB than above.

30
Ownership and Incentives

• So better that 2 own the asset, although he has
  no decision to motivate, because if 1 owns, then
  incentives unbalanced and e1 too high, driving
  down e2. Better balanced but weak incentives.
• Perhaps better that third party owns so that
  excessively strong incentives removed (cf GHM)

31
Implication

• Must think of incentives systemically, consider
  interactions.
• Next intertemporal aspects.