52
3 Economic Framework of Conservation
Landraces have provided many individual traits which have been introduced into
existing improved breeding lines. In this context, Duvick is cited by Salhuana and
Smith (1996) that the number of landraces utilized in the five most important
maize inbreds are increasing over time: while in 1930,5 landraces were utilized, it
was already 11 in 1960 and went up to 27 in 1990. Evenson (1994), Evenson and
Gollin (1996), and Gollin (1996) show with a breeding production function study
on rare traits that 10% of the overall gain in rice productivity in South Asia can be
attributed to the size and evaluation status of the collections of landraces and
related populations used by breeders. This indicates a gain from the material in
these collections of about 0.2%, or about US $ 150 - 200 million per year. This
value is based on another estimate by Evenson and David (1993). It was estimated
that the average value for each of the 1,400 varieties of rice released in India,
Pakistan, Bangladesh, Philippines, Thailand, Indonesia, and Brazil in 1990 was
US $2.5 million per year. These figures have to be sensitively interpreted, because
they represent a gross value before the deduction of costs. Furthermore, as with all
other estimates as well, it does not give the isolated value of the genetic material
used, but rather the aggregate value of both the genetic resources, as well as the
contribution through research inputs as capital, labor and technology involved in
plant breeding (NRC, 1993).
Other estimates have been made on the value of CIMMYT-based wheat
germplasm to agriculture in OECD countries. These range from $300 million to
$11,000 million per year (e.g., Mooney, 1993). The large differences in the
estimates reveal the inaccuracy in assessing their value. However, other estimates
based on a simple search model show that genetic resources for food and
agriculture are not scarce, therefore not of much economic value (Simpson and
Sedjo, 1996). These examples show that an economic analysis of the value of
PGRFA is still in its infancy. Concentration solely on the consumptive or use
value of PGRFA and further research is required to provide better estimates.
Generally speaking, it is assumed that the loss of PGRFA diversity implies the
lost chance for the utilization of a specific trait or genetically coded information in
crop improvement. The outbreaks of diseases or pests such as the Irish potato
famine, which could be given a check by incorporating needed traits into existing
varieties serve as justification. But, as Sen has shown (1981), crop failure does not
exclusively correspond to famine. It is rather a question of other policy,
institutional, and market failures, partly associated with war, violence, or
deliberate exploitation. Resistance breeding and utilizing more than only one
potato variety were factors rectifying the famine, but not the most important and
immediate ones. Taking into account the time lag of 5 or more years between the
start of breeding as reaction to a sudden outbreak and the supply of a new variety
especially does not verify the assumption that crop improvement through
genetically coded information from landraces is the source of insurance against
crop losses, as Brown and Goldstein (1984) implied.
Following the estimates of Evenson and Gollin, economic considerations
predict an under-investment in the conservation of PGRFA. But the costs of
extinction can be overestimated if the opportunities to find substitutes for crop
losses are not recognized. On the one hand, crop losses similar to the Irish potato
3 Economic Framework of Conservation
Landraces have provided many individual traits which have been introduced into
existing improved breeding lines. In this context, Duvick is cited by Salhuana and
Smith (1996) that the number of landraces utilized in the five most important
maize inbreds are increasing over time: while in 1930,5 landraces were utilized, it
was already 11 in 1960 and went up to 27 in 1990. Evenson (1994), Evenson and
Gollin (1996), and Gollin (1996) show with a breeding production function study
on rare traits that 10% of the overall gain in rice productivity in South Asia can be
attributed to the size and evaluation status of the collections of landraces and
related populations used by breeders. This indicates a gain from the material in
these collections of about 0.2%, or about US $ 150 - 200 million per year. This
value is based on another estimate by Evenson and David (1993). It was estimated
that the average value for each of the 1,400 varieties of rice released in India,
Pakistan, Bangladesh, Philippines, Thailand, Indonesia, and Brazil in 1990 was
US $2.5 million per year. These figures have to be sensitively interpreted, because
they represent a gross value before the deduction of costs. Furthermore, as with all
other estimates as well, it does not give the isolated value of the genetic material
used, but rather the aggregate value of both the genetic resources, as well as the
contribution through research inputs as capital, labor and technology involved in
plant breeding (NRC, 1993).
Other estimates have been made on the value of CIMMYT-based wheat
germplasm to agriculture in OECD countries. These range from $300 million to
$11,000 million per year (e.g., Mooney, 1993). The large differences in the
estimates reveal the inaccuracy in assessing their value. However, other estimates
based on a simple search model show that genetic resources for food and
agriculture are not scarce, therefore not of much economic value (Simpson and
Sedjo, 1996). These examples show that an economic analysis of the value of
PGRFA is still in its infancy. Concentration solely on the consumptive or use
value of PGRFA and further research is required to provide better estimates.
Generally speaking, it is assumed that the loss of PGRFA diversity implies the
lost chance for the utilization of a specific trait or genetically coded information in
crop improvement. The outbreaks of diseases or pests such as the Irish potato
famine, which could be given a check by incorporating needed traits into existing
varieties serve as justification. But, as Sen has shown (1981), crop failure does not
exclusively correspond to famine. It is rather a question of other policy,
institutional, and market failures, partly associated with war, violence, or
deliberate exploitation. Resistance breeding and utilizing more than only one
potato variety were factors rectifying the famine, but not the most important and
immediate ones. Taking into account the time lag of 5 or more years between the
start of breeding as reaction to a sudden outbreak and the supply of a new variety
especially does not verify the assumption that crop improvement through
genetically coded information from landraces is the source of insurance against
crop losses, as Brown and Goldstein (1984) implied.
Following the estimates of Evenson and Gollin, economic considerations
predict an under-investment in the conservation of PGRFA. But the costs of
extinction can be overestimated if the opportunities to find substitutes for crop
losses are not recognized. On the one hand, crop losses similar to the Irish potato