Kathleen Segerson
chosen voluntarily, then the direct market benefits of meeting the standard
should be the same regardless of whether the standard was met voluntarily.
In other words, the quality of the product will be identical and hence the
direct consumer benefits (as reflected in willingness-to-pay) will be
identical. However, the firm is not likely to receive much in the way of
public relations benefits (e.g., good will) from meeting a mandatory
standard. Since these public relations benefits are included in Bv, we
would expect Bv≥Bm. Similarly, since Cv represents the minimum cost of
meeting the standard, Cm cannot be less than Cv, i.e., Cm≥Cv. The actual
magnitude of Cm will depend on the nature of the mandatory controls, as
well as the transactions costs associated with meeting the mandatory
standard. If those controls specify not only a performance standard but
also a process or design standard, i.e., they also specify how the
performance standard is to be met, then it is possible that the standard will
not be met in a least cost way.22 Conversely, if the mandatory controls
simply take the form of a performance standard, they allow the firm
maximum flexibility in deciding how to meet the standard. In this case,
the firm would still be free to choose the least cost method. Even in this
case, however, if compliance with the mandatory controls generates
significant transactions costs, we would expect Cm>Cv.
If the firm does not adopt the measures voluntarily and the government
does not impose a mandatory standard, then no protection measures are put
in place. As a result, the probability of contamination is higher than it
would have been with the standards. Let p denote the probability of
contamination without controls, where p>q. Without controls, the firm's
payoff is Bo-L if contamination occurs and simply Bo if it does not. In this
case, Bo reflects the revenue from the sale of the product, given the higher
probability of contamination.
A comparison of the expected payoffs from the tree depicted in Figure 1
suggests that the firm will choose to undertake the protective measures
necessary to meet the given standard voluntarily if and only if23
(1) Bv+S-Cv-qL ≥ r[Bm-Cm-qL] + (1-r)[Bo-pL].
To understand the implications of (1), we consider some special cases.
First, suppose that the government plays no role in promoting food safety.
In other words, it offers no subsidies and there is no threat of direct
regulation, implying S=r=0. In this case, (1) reduces to
22 Food safety controls have historically focused primarily on process-related
requirements. Recently, however, there has been a shift in the nature of some
regulation, toward granting more flexibility to firms. See Caswell and Henson
(1997), Caswell and Hooker (1996), and Unnevehr and Jensen (1996) for related
discussions.
23 This assumes that firms are risk neutral, or that they are able to diversify risks
through the purchase of insurance or other risk spreading mechanisms.