Reputations, Market Structure, and the Choice of Quality Assurance Systems in the Food Industry

is discovered as falsely certifying a product by assuming that the firm loses its reputation,
which forces it out of the certified market.³ Note however that the “reputable” processor
obtains the price for the certified commodity no matter what the actual quality might be,
because customers cannot assert a priori whether the claims made by the processor are false. In
other words, processors will be trusted until proven wrong. Consumers’ trust is what defines
the states of the world in this model. For a given processor, demand is state contingent, where
the states of the world reflect whether it is trusted by consumers or not. For this sort of
punishment mechanism to have an impact on a firm’s decisions, modeling more than one
period is required (Klein and Leffler).

We introduce ω∈[0,1] to measure the degree to which consumers can ascertain the
actual quality of the good, where ω= 1 implies that quality is perfectly observable after
consumption, or the sought-after characteristic is an experience attribute. Credence attributes
are represented by ω= 0 . Intermediate values of ω may be interpreted as attributes that are
occasionally detected, or detected only by a proportion of consumers. For example, only a
proportion of consumers will be able to discern whether a steak was produced from cattle fed
exclusively grain.

We are now in a position to examine how fundamental characteristics of the economic
environment influence decisions about the implementation of a QAS paying special attention
to market structure and different forms of punishment. Clearly, QASs will be observed if
E_(∏i,r₍

s,y))≥πⁱ=0 for some s∈ S, and y > 0, where πⁱ is the profit level available in an
alternative market. That is, if there is a combination of output rate and QAS that makes the
expected return of the value-added market positive, then the firm has an incentive to adopt the
QAS and supply the high-price market. Throughout the analysis, we assume for mathematical

<<   4   5   6   7   8   9   10   >>

More intriguing information