Lecture 1: Introduction to the political economy of natural resources Lecture 2: Theories of collective action, cooperation, and common property

Property Rights and Collective Action in Natural Resources with Application to Mexico

• Lecture 1: Introduction to the political economy of natural resources

• Lecture 2: Theories of collective action, cooperation, and common property

• Lecture 3: Principal-agent analysis and institutional organization

• Lecture 4: Incomplete contracts with application to Mexico

• Lecture 5: A political economy model

• Lecture 6: Power and the distribution of benefits with application to Mexico

• Lecture 7: Problems with empirical measurement with application to Mexico

• Lecture 8: Beyond economics: An interdisciplinary perspective

Why property rights important

• Allow markets to function

• Increase incentives to invest

• If transaction costs = 0, fully defining property rights gets us to an efficient outcome

Transaction costs

• If transaction costs > 0, asset ownership confers power

• Transaction costs of an exchange:

• Negotiating the agreement
• Writing a contract
• Observing/monitoring the agreement
• Enforcing agreement of contract is broken

What is a property right?

• Residual rights of control

• Ability to specify unspecified uses

• Ability to alienate others from use of the property

• Who owns what determines efficiency if transaction costs > 0

• Incomplete contract (IC) theory: concerned with the most efficient allocation of property rights.

What is an incomplete contract?

• Transaction costs make contracts incomplete

• Bounded rationality:

• Background assumption to incomplete contracts.
• Cannot think through all branches of a decision tree.
• Cannot represent everything in a contract.
• E.g. complicated bilateral relations.

• People deal with this bounded rationality in a substantively rational way.

Incomplete Contract Theory

• Grossman and Hart (1986)

• Hart and Moore (1990)

• Hart (1995)

• Different than Williamsonian transaction costs analysis

• Allows renegotiation

Model

• 2 parties: i = A and B

• Each party may be an “expert” to work with some physical asset KA and KB respectively

• Both parties A and B can expend investments (a and b, respectively) in human capital

• These investments are nonverifiable

• Other party cannot buy this capital

Timeline of Production Relationship

Model Notation

• Let K = {KA, KB}

• Ki is set of asset owned by i

• Denote KA  KB = K

• Denote KA  KB = 

Default payoffs

• Individual payoffs if do NOT work together:

• VA[KA, a, 0]  VA[KA,a]

• VB[KB, 0, b]  VB[KB,b]

Payoffs with Trade

• Total payoffs if work together:

• V[K, a, b]

• Payoff to us: less notation!

Physical Asset versus Human Capital (or Relationship) Specificity

• Investment ai party i is:

• Nonspecific when the marginal return does not depend on Kj nor aj

• Purely relationship specific when the marginal return is 0 unless parties work together

• Relationship specific when the marginal return positively depends on access to aj

• Purely asset specific if the marginal return depends (only) on party i’s access to Kj

Basic Assumptions

• V[K, a, b] > VA[KA,a] + VB[KB,b]  Ki

• GAINS TO TRADE: it is efficient to work together
• Let ai = {a if i=A, b if i=B}

• Vai [.]  Viai [ K, ai]  Viai [ Ki, ai]  Vai [Kj, ai]  Vai [, ai]

• Marginal returns from investments are weakly larger the more human and physical capital a party has access to, i.e. relationship specificity exists.

Asset complementarity

• An asset is complementary when:

• Viai [K, ai] > Viai [Ki, ai]

• (access to both assets K makes you more productive)

• An asset is strictly complementary when:

• Viai [Ki, ai] goes to 0

• (you need access to Kj as well)

• An asset is independent when:

• Viai [K, ai] = Viai [Ki, ai]

• (one asset suffices for your productivity)

Equilibrium analysis

• Backwards induction: start with Date 2

• Date 2: Investments a and b are “sunk”

• Since V(.) > VA (.) + VB(.)  a,b,Ki, then parties will agree on “working together” but choose investments noncooperatively  ex post efficiency

• Division of joint surplus V(.): Nash bargaining solution

• Party i obtains:

• Default payoff: Vi[Ki , ai]
• Fraction i  [0,1] of “excess surplus”:

• V = V(.) - VA (.) - VB(.)

Who should own what?

• Compare investments levels under various ownership configurations and contracting conditions (e.g. specificity, complementarity, cooperative v. noncooperative investments)

• The configuration yielding highest total surplus (both A and B’s payoffs combined) is most efficient given the conditions.

A owns KA and KB: Integration

• UIA = VA[K, a] + A[V[K,a,b] - VA[KA,a] -VB[,b] ] - a

• UIB = VB[0, b] + (1-A)[V[K,a,b] - VA[KA,a] -VB[,b] ] – b

• FOCs:

• A: (1- A) VAa [ K, a] + A Va [ K, a, b] = 1  aI

• B: A VBb [,b] + (1- A) Vb [ K, a, b] = 1 bI

A owns KA, B owns KB: Nonintegration

• UNIA = VA[KA, a] + A[V[K,a,b] - VA[KA,a] -VB[KB,b] ] - a

• UNIB = VB[KB, b] + (1-A)[V[K,a,b] - VA[KA,a] -VB[KB,b] ] – b

• FOCs:

• A: (1- A) VAa [KA, a] + A Va [ K, a, b] = 1  aNI

• B: A VBb [KB,b] + (1- A) Vb [ K, a, b] = 1 bNI

First-Best Efficient Outcomes

• Parties “work together” and choose investments cooperatively  first-best outcome

• Investment levels are:

• [a*, b*] = argmaxa,b V{K,a,b] – a – b

• FOCs: Va = 1 ; Vb = 1

• Simple!

First Result

• Under any ownership structure (i.e. integration or nonintegration) there is underinvestment in relationship-specific investments (Hart 1995), that is:

• a* > aI or aNI

• and

• b* > bNI or bI

Marginal productivity of investments

Other Results

• If investments by one party are inelastic, then integration by the other party is optimal

• If assets are independent, then nonintegration is optimal

• If assets are complementary, then some form of integration is desirable.

• If one party’s HK is essential, then integration by that party is optimal

• If both parties’ HK are essential, then all ownership types are equally optimal.

Critique of Incomplete Contracts

• Maskin and Tirole (1999):

• Agents have bounded rationality but deal with noncontractibility in a rational way
• does BR constrain set of feasible contracting outcomes relative to complete contracts?
• Counter-theorem: Nondescribability is irrelevant under certain conditions

• Segal and Hart: as number of possible states of world increase, null (incomplete) contract is optimal.

Application to Mexico Community Forestry Production

• What explains the vertical integration pattern across Mexican communities with forest?

• Apply and extend incomplete contract model

• Assume two parties:

• A = community
• B = downstream firm/buyer

• Assets:

• Forest land, extraction equipment, milling and processing equipment.

Assets and investments: Forest management

Assets and investments: Extraction

Assets and investments: Milling

Assets and investments: Secondary processing

Examples of Unforeseeables

• Rainy season

• Access problems

• Equipment failure

• Scheduling problems

• Timber condition

• Failure to pay, deliver

• Un-monitored extraction

An IC Model

• Survey data from 43 observations in Oaxaca (see Antinori (2000))

• Ownership possibilities:

• Assume communities can own forest land and any downstream equipment
• Assume outside private firms can own any downstream equipment for wood production and can contract with communities for raw material.

• Tradeoffs: need for specialized expertise and overcoming collective action hurdles v. benefits of ownership (see Antinori and Rausser (2007) for details)

Econometric Model of Vertical Integration

• Ordered logit (see paper for details)

• Dependent variable: level of community vertical integration (sale of stumpage, roundwood, lumber of secondary goods)

• Independent variables:

• Influences on ability to coordinate, produce:

• Past mechanical training (+ and significant)
• Initial logging roads (not significant)
• Parastatal existence (+ and significant)

• Contracting costs, valuation:

• Forested hectares (+ and significant)
• Forest quality (initial) (+ and significant)
• Distance to pop center (+ and significant)
• Non-commercial timber forest uses (not significant)

Conclusions

• Ownership of assets critical for control over benefits when transaction costs present

• Limitations of PR approach:

• Collective decisionmaking still not completely modeled in firm.
• Beyond property rights as control over assets

Asset Wealth

• Oliver and Shapiro: Black Wealth/White Wealth:

• Persistent wealth gap versus income gap in the United States between black and white
• Limits opportunities