can be optimal. However, when this is the case the budget does not allow
any change in education policy. We then consider the case in which a foreign
loan is available and find that the country can benefit from a foreign loan as
the welfare gains resulted from relaxing the government budget constraint
exceed the welfare losses due to the lump-sum taxation imposed in order
to repay the loan.9 We also find that reversals in the opposite direction,
moving down the skills chain, can also be optimal and that such reversals
are not budget restrained.
In the following section we take a preliminary look at the data, in section
3 we build our theoretical model, in section 4 we derive the results for the
autarky case and in section 5 we open the economy to international trade
and in the last section we offer some final comments.
2 A Preliminary Look at the Data
In the following five tables we present data on educational attainment and
sectoral exports for a sample of 24 countries for the years 1980, 1990 and
2000.10 For educational attainment we show the proportion of the popu-
lation aged above 25 with secondary education and the corresponding pro-
portion for those with tertiary education. The export data indicate the
percentage of exports that are high tech goods. We have used two thresh-
olds to separate high-tech from low-tech sectors. The Broad definition of
high-tech goods sectors uses a lower threshold and hence delivers uniformly
higher numbers than the Narrow definition.
Table 1a includes countries which, like the United States, have been
exporting a large proportion of high-tech products throughout the sample
period.11 These countries also show very high educational attainment levels
9The theoretical literature on dynamic comparative advantage suggests that if the ini-
tial gains of trade are sufficiently high and thus relax the factors constraining growth (in
our case the limited budget constraint) then a reversal of exports from low-skill goods to
high-skill goods might become optimal. This suggests that economies that move up the
chain must be economies that grow fast.In this paper we focus on distributional aspects
of government policy and, for analytical tractability, we abstract from dynamic consid-
erations. For theoretical work on dynamic comparative advantage see Bond, Trask and
Wang (2003), Redding (1999) and Ventura (1997).
10The sources of the data are the same as those used for figures 1 and 2. The sample
comprises of countries for which all data were available. The only exceptions are Bulgaria
and Romania for which export data for 1980 are not available but were included given
that their particular group is small.
11It is convenient to use the broad definition in our ananlysis where a proportion above
fifty percent is interpreted as the country having a comparative advantage in high-tech