Since 2000, Korea has experienced a sudden downturn in the amount of FDI
(Foreign Direct Investment) inflow which has continued ever since. Rather than
this being unique to Korea, the declining trend in the size of FDI is very much
a global phenomena. Hence, global competition to attract FDI among governments
has intensified even more. Under the policy competition for the inducement of
FDI, lies the belief that the positive benefits of FDI on the domestic economy
as a whole exceeds the overall costs of various subsidiary measures. It is natural,
therefore, for economiststo investigate scientific evidences on the existence
and content of benefits associated with the inflow of FDI, on the proposition
that the positive externalities are large enough to call for government intervention,
etc.

In general, FDI is known to be advantageous to the targeted country in various
ways. For example, improving the balance of payments, increasing production
and employment, raising productivity of the target company as well as that of
other companies in the same or related industries. While economists agree on
most, they have not reached an agreement on the issue of productivity spill-over.
Since the beneficiaries of productivity spilloversare usually different from
the direct stake-holders of the FDI, the productivity spillover effect can be
classified as one of the positive externalities of FDI. It is crucial, therefore,
to study this productivity spillover to correctly measure the economic value
of FDI, and to verify the validity of governmental support for FDI.

There exist several studies in Korea on the direct effect or indirect intra-industry
productivity spillover effect of FDI. Despite its importance, the indirect inter-industry
productivity spillover effect of FDI has never been analyzed in Korea. Given
the circumstances, this study examines the effect of FDI on the productivity
of firms in the same industry as well as firms in upstream or downstream industries.
This study uses panel data set of listed manufacturing companies in the Korea
Stock Exchange and KOSDAQ as well as companies subject to external auditsduring
1991 and 2000.

First, as indicated by the results, the dummy for foreign direct invested
firms had a coefficient that was positive and statistically significant. Therefore,
it can be concluded that the productivity of firms with FDIs is higher than
that of non FDI domestic firms with other things being equal. When the results
obtained by applying the treatment effect model to correct for the endogeneity
bias are compared to the ordinary least squares estimates, it can be shown that
the coefficient of the foreign dummy in the case of treatment effect estimation
is bigger than that of the ordinary least squares estimation. It can be concluded,
therefore, that the direct effect of FDI on productivity is not a result from
endogeneity, and that ordinary least squares methodology underestimates the
magnitude of this direct effect.

Next, when the intra- and inter-industry productivity spillovers are examined,
the coefficients of intra-industry and downstream industry FDI share are both
found to be positive and significant. The backward productivity spillovers are
especially bigger and more significant among all the variables, and are not
affected by the model specification and share calculation method employed. The
forward productivity spillover, on the contrary, appeared to be unstable depending
on the model specification and share calculation method. In some cases, it appeared
to be negative and significant. It can be concluded, therefore, that FDI has
a positive productivity spillover effect on the firms in the same industry as
well as firms in the upstream industry.

This study may be considered valuable since it is dealing with inter-industry
productivity spillover effects of FDI which has never been studied before in
Korea. The existence of such an effect provides evidence on the public good
aspects of FDI and thereby can be a rationale for governmental support for attracting
FDI. It can also be suggested that those companies that are closer to the consumers
in the value chain, i.e., companies producing final finished products are the
ones that should be the focus of governmental support to maximize the overall
benefits from FDI.

As in other cases, this study does have its shortcomings. For instance, the
analysis does not provide answers to such questions concerning the optimal amount
of subsidy and optimal choice of subsidiary measure. The study cannot separate
the enhancement of productivity due to the transfer of technology and increased
competition. The study, also, relies on high level industry classification.
For the future, overcoming these problems should be a worthwhile challenge to
those who are able.