ARE 290 Spring 2006

1
ARE 290 Spring 2006 Economic and Environmental
Tradeoffs in Agricultural Systems Integrating
Bio-Physical and Economic Models

• Assignment 10 discussion
• Model calibration validation
• Carbon sequestration implementation in MK model

2
ARE 290 Spring 2006 Economic and Environmental
Tradeoffs in Agricultural Systems Integrating
Bio-Physical and Economic Models
3
(No Transcript)
4
Homework 10 Discussion

• Note sensitivity of System 0 to INP1
• Changing model parameters sensitivity of
  intercept useful to re-estimate model with
  parameter restriction imposed so model predicts
  correctly in the base case.

5
Model Calibration

• Model equations should predict sample means if
  model is correctly formulated
• But land allocation is based on assumption of
  expected returns needs to be validated
• Models represent population means, not generally
  equal to individual expectations which should
  vary in the population
• E.g., V f(x) u is population model,
  calibrated model to represent variation in
  individual expectations could be V f(x) ?u, 0
  lt ? lt 1, so that variability of expected V is
  less than variability in the population.
• Similarly, could define expected price as P
  g(z) ?w if price expectations vary less than
  actual prices.
• Also may adjust expected prices to reflect.
• Calibration can be considered analogous to
  estimation this concept could be formalized
  (research topic!)

6
Model Validation

• But land allocation is based on assumption of
  expected returns needs to be validated
• Compare mean land allocation to simulated values
  mean of binomial distribution is sufficient
  statistic.

7
Note on Modeling Log-Normal Expected Values

When output is modeled as V f(x)eu be aware
that the mean of V is E(V) f(x) exp(?2/2),
where Var(u) ?2 In other words, when
simulating expected V, the correct value is as
above. Note that since exp(?2/2) gt 1, f(x) is an
underestimate of E(V).
8
Carbon Sequestration in MK Model

• Supplemental reading www.tradeoffs.montana.edu/pu
  blications (carbon sequestration and poverty)
• Contract design
• Market imperfections constrain input use
• Contract requires fertilizer and OM input and
  makes them available at market price
• C payments not likely to be sufficient to cover
  cost of fertilizer, e.g., in Machakos 60 kgN
  600 kgOM gives carbon rate of 0.3 MgC/ha/season
  carbon price50/MgC gives 15/season, or 38 kgN.
• Access to OM is likely to be a binding
  constraint, may lead to endogenous price.
• Verification changes in land use can be verified
  at low cost, but changes in input use may be
  difficult. Will need low-cost enforcement
  mechanisms (e.g., self-enforcement) or risk of
  default likely to be high as in financial
  markets.
• Risk
• Increasing N may increase production risk, but OM
  likely to reduce it (do data support this?)
• Other soil conservation practices (terracing)
  likely to reduce risk
• PES less risky than agricultural returns.

9
Carbon and Poverty

• Scenario implementation
• C rates for partial adopters for those using
  zero nutrient inputs the carbon rate is C0. For
  those using a positive amount less than the
  contract rate xC, we assume the rate is
• C (xC xB)/ xC
• -- C rate with multiple inputs Ci Si C0, i
  fert, manure.
• -- Contract duration make contract participation
  decision in cycle 1, must remain in contract for
  duration (permanence problem)
• -- note incorporation of contract rates of input
  in output equations
• -- note opportunty cost of residue incorporation
  must be included in expected returns under
  contracts.

10
Simulated participation in carbon contracts,
Machakos, Kenya
11
Simulated participation in carbon contracts,
Senegal Peanut Basin (R denotes percent of crop
residue incorporation, TC denotes transaction
cost in dollars per hectare per season)
12
Rate of change in soil carbon versus poverty gap
with carbon contracts, Machakos, Kenya (Left-most
point corresponds to a zero carbon price, the
price increases to 200/MgC at the right-most
point)
13
Rate of change in soil carbon versus poverty gap
with carbon contracts, Senegal peanut basin
(Left-most point corresponds to a zero carbon
price, the price increases to 200/MgC at the
right-most point R denotes percent of crop
residue incorporation required in the carbon
contract)
14
Rate of change in soil carbon versus poverty gap
with carbon contracts, Senegal peanut basin
(Left-most point corresponds to a zero carbon
price, the price increases to 200/MgC at the
right-most point R denotes percent of crop
residue incorporation required in the carbon
contract)
15
JP and GME estimates of mean production
elasticities for maize production in Machakos,
Kenya