Budget 1992). Whatever rate one agrees upon, such increase in R&D costs affects all R&D projects
and hence shifts the R&D opportunity curve down. However, to our knowledge none of the reported
rate-of-return studies in our sample have actually included an excess burden rate in its calculations.
This means that, collectively, R&D opportunity curves are portrayed too optimistically.
Although often seen as a rather typical US, anti-government sentiment, the influence of the
literature on excess burden on taxation policies has been quite significant worldwide. Very few
countries, however, actually use an excess-burden rate in their evaluations of government programs.
Nevertheless, it does not mean that it is not a factor of importance in making a decision whether or not
to fund a government program. Politicians want to be reelected and they know that they do not make
themselves popular by raising taxes unless they can show that the programs that are being funded
bring significant (economic) benefits to society. The excess-burden rate makes this economically
rational resistance to taxation explicit and provides an explanation of why, when no provision is made
for the excess burden factor, politicians apparently stop funding agricultural R&D projects (or for that
matter any other government project) before the marginal project hits the social cutoff rate.
As already discussed in section 3, the uncertain and risky nature of the economic potential of
an R&D investment is another dimension that may very well affect the level of investment in
agricultural R&D. Discounting for this risk and uncertainty will shift the position of the ex ante R&D
opportunity curve down. Ex post, this expresses itself in an underinvestment gap.
Even when full and accurate information could be assumed, the selection of R&D projects is
influenced by many other factors that skew it away from the economic optimum. Stakeholder
participation and political lobbying, for example, contribute positively by bringing more accurate and
relevant information to the selection process of R&D projects and the setting of an R&D agenda. Its
downside, however, is that the outcome of this process may not necessarily be balanced. De Janvry et
al. (1989), for example, argue that in a typical dual-economy setting rich farmers, because of their
better political lobby capabilities, skew the selection of R&D projects into their favor. In such
instances, transparent priority setting procedures that manage to balance the various political and
economic interests are needed to enhance the economic rationality in the selection of R&D projects
and hence reduce the underinvestment gap. Similarly, de Gorter and Zilberman (1990) focus on how
22