Cross-Commodity Perspective on Contracting: Evidenc e from Mississippi
A higher value of ® has a greater cost reducing consequence for asset specificity: Ck® < 0
and CX® < 0.
Assume that governance costs are given by:
Gi = β + V(k); and
Gm = W(k)
where β > 0 is the bureaucratic cost parameter and Vk ≥ 0 and Wk > 0. Further, it is
assumed that Wk > Vk for all common k, or the marginal cost of governance with respect to
asset specificity is greater for spot markets than for any type of contract. It is clear from
this that output will be chosen to equalize marginal revenue with marginal production cost,
while optimal asset specificity will be chosen to minimize joint pro duction and governance
costs.
Examining profit as a function of asset specificity yields a central hypothesis (Figure
1). It is assumed that the level of output chosen is optimal for each level of asset specificity.
There is only one profit function for spot sales, but a family of profit relations for contracts
of different types depending on the bureaucratic cost parameter β. The choice of mode
of production depends on the profit relation with the highest peak. For example, with a
relatively low bureaucratic cost, β1, contracting would be chosen. By contrast, spot sales
would be chosen when the bureaucratic cost parameter, β2 , is relatively high. Clearly, then,
it is the relative transactions cost between spot sales and contracting that determines the
mode of production. Thus, imp ortant variables to examine are variables that relate to