∏w-∏x = Γ(r,τ√) ≡ γ(r∕) + χ(r,τ)
+ ” (8)
where: χ(t,τ) ≡ (n-l)[π(τ)-π(r)]
+ -
Now the full gains from FDI, denoted by Γ(t,τ,f ), come from two sources. There is a
potential gain to tariff-jumping into an individual market, γ(t,f ), as before. In addition, the
reduction in internal barriers yields an extra source of return to FDI, denoted by χ(t,τ). A
plant located anywhere in the union now benefits from preferential access to the other n-1
markets, and so can serve as an "export platform". This gain is always positive, since π(t)
is decreasing in t. Hence, FDI may be preferable to exporting from outside the union even
when the gain to tariff jumping into a single union country, γ(t,f ), is zero or negative.
Similarly, the profits from establishing plants in m union countries, ΠFm, are now:
Πf"' = rn[π(O) ~9) + (ra-m)π(τ) (9)
= ∏f,m^1 + γ(τ5∕)
Compare this with (4) and (5). Once a single plant has been established in the union, it is
less profitable to establish additional plants: lower internal tariffs mean lower gains from
internal-tariff-jumping. Since we have already seen in (8) that the relative profitability of
exporting has fallen, we can conclude that it may now be profitable to have only one plant.
However, it is never profitable to have more than one and less than n plants. Since all union
countries are identical and all intra-union tariffs are the same, there is nothing to be gained
from multiple union plants short of one in each country. The profits of the latter, ΠFn, are
still given by (6).
The relative attractiveness of FDI with a single union plant rises as internal tariffs are
progressively reduced. Differentiating (7):