To account for the crosscountry differences in
learning capabilities and labor skills, we use the variables
EDZND and ED3RD. They measure the share of
the appropriate age-group in secondary and third level
education in each host country 1980, and are taken
from UNESCO’s Statistical Yearbook 1990, Table
3.2. We expect both variables to have a positive intluence
on the affiliates’ technology imports, because the
marginal technology transfer costs are lower when the
level of education is higher. It should already be noted
here, however, that the small variance in ED2ND is
likely to reduce its observed significance.