bargaining power for greater FDI inflows are those with high βh. MNEs then
rationally expect the “MAI club” to be formed by countries that are, on average,
more “greedy” than others. Doing business there would be profitable only if
MAI members commit to self-restrain their bargaining power vis-à-vis MNEs
sufficiently. We also see that the realization of MAI would be facilitated when
βl is high, namely, in a world where, on average, the balance of power is in favor
of countries, and not of MNEs. The leverage of a MAI agreement in attracting
FDIs will be higher under this scenario. At intermediate levels of ° and βl the
extent of coordination failures is particularly acute. There, results depend a
lot on expectations and “political climate”. If there is widespread presumption
that a MAI agreement will be implemented, this expectation will be fulfilled; if
expectations are pessimistic, no agreement will be reached.
We did not discuss so far the determination of ° , namely the strictness of
MAI. Parameter ° in the analysis has been taken as exogenous. This corre-
sponds to a situation where the design of the investment agreement is decided
independently by a third party, for instance, an international agency, and pro-
posed to countries, that independently may decide to participate or not. We
know that high ° is needed for the realization of MAI. Hence, if an international
agency can independently and optimally choose °, it will choose a rather “se-
vere” agreement. However, in many real world instances the determination of
° requires the consensus of countries, for instance through a voting procedure.
If this is the case, the implementation of MAI may be hard to achieve without
some form of side-payments to MAI losers (obtainable, for instance, by linking
the negotiations of MAI with those on some other issue).26 The reason is that
some countries might anticipate that they will lose from the implementation of
MAI, and will consequently oppose a strict MAI (high °). In the absence of
side payments, the realization of an agreement on FDIs will be easy only in a
world where the distribution of MNEs is relatively equal across countries. How-
ever, in the light of the results in Proposition 5, we see that things are different
in the case of bilateral or regional investments treaties. When the agreement
is ex-ante restricted to a subset of countries, we can expect a relatively easy
implementation of preferential investment agreements. Also the countries with
few or no MNEs’ holdings might anticipate to gain from the agreement because
of more abundant FDIs inflows.27 This might explain why we have observed
in recent times a strong opposition of some countries to the implementation
of a world-wide MAI together with the proliferation of bilateral and regional
agreements on international investment.
26See also Hoekman and Saggi (1999) for arguments in favor to linking multilateral invest-
ment agreements with negotiations on other issues in a ”grand bargain”.
2 7 Bilateral investment treaties b etween develop ed and developing countries are indeed quite
common. Morevoer, as in the case of Nafta, regional investment agreements may involve
countries at different stages of development.
20