12
amount of fraud is reduced. 6 In our paper, firms would have incentives to reduce the number
of mistakes they make as consumers become increasingly able to discern qualities.
There exists a key trade-off between the benefits and costs of information acquisition on the
part of processors. Having a more precise QAS, though costly, decreases the probability that
firms will lose consumers’ trust. Furthermore, as the expected losses derived from consumer
distrust increase, the payoff from the processor becoming better informed about actual quality
increases.
Using Cramer’s rule to solve for
∂y*
-, we find that the sign is ambiguous without imposing
∂ω
further structure, since ω enters by itself in equation (4). Moreover, system (5) (and the sign
just uncovered) tells us
thatsgn j''ʌ *∕ωl ) =sgn (d π∂, ∂s )
which is not implied by the
maximization hypothesis alone. Since it is reasonable to assume that raising the levels of
∂2π ∂2C
controls increases the marginal costs of production, and noting that----=--, we expect
∂y∂s ∂y∂s
the optimal output rate to decrease as ω increases.
The question of how the value that producers place on the future affects the optimal choices of
QAS and output levels can be explored through a similar exercise. The results of
differentiating equations (3) and (4) with respect to β and using Cramer’s rule (again omitting
arguments) are as follows:
∂s*
∂β
=H -1
∂2π
∂y∂s
∂2π
H is the Hessian matrix shown in (5). Thus we
know that
where
∂2π(y, s; a)
= sgn -----------
^ ∂y ∂s ^
and ---≥ 0 . As the future becomes more important, it is
∂β
more valuable for processors to invest in quality assurance systems that give them a longer
∂y*
expected presence in the market. The sign of is ambiguous as before (and due to the
exact same reasons). However, the previous discussion suggests it is negative. Increasing the