Here is an e-mail from a well-informed guy who occasionally sends me stuff on tax and trade issues. Bottom line: It’s really complicated. E-mail:

Let’s set aside for a moment the fact that all this stuff is just very
confusing even to those of us who have attempted to study it. I’ll try
to pick through all this. This is another don’t use my name (whatever
the going acronym is for that) e-mail.
Interacting with the deferral is the following: U.S. tax law taxes the
global income of multinationals, but also allows a credit of foreign
income taxes against the U.S. corporate income tax. So for U.S.
affiliates in places where the corporate tax rate is higher than in the
United States (like much of Europe) the effective tax rate they pay is
the foreign tax rate, and the foreign government gets it all. An
accountant may know this better than me, but I don’t think repealing
deferral changes the firm’s tax burden or the firms’ behavior.
Seemingly repealing deferral would have more impact on the propensity of
U.S. firms to invest in low-tax countries.
Now here is the problem – for a lot of firms, pre-existing foreign
direct investment (FDI) is part of doing business. About 2/3 of U.S.
FDI is directed at serving host-country markets, not in selling back to
the U.S. So there’s a big incentive to keep those investments
regardless of taxes. But repealing deferral hits the main way in which
FDI is financed. You can either transfer funds from the U.S. parent to
the affiliate, have the affiliate borrow in local or world capital
markets, or reinvest the profits of the affiliate. The biggest chunk of
U.S. *outbound* FDI is these reinvested profits, in other words letting
all your old foreign affiliates mostly self-finance through retained
earnings.
So if you’ve got to have the affiliate there, and it becomes really
inconvenient from a tax standpoint to finance its capital expansion
through retained earnings, then your typical corporate treasurer would
switch to one of the other two methods. Either have the head office
borrow on U.S. markets, or have the overseas affiliate borrow on U.S.,
local, or world markets. Now if it matters at all, then this would tend
to push interest rates up and crowd out Americans wanting mortgages,
college tuition loans, or whatever. If it doesn’t matter a lot, then
the Treasury wouldn’t really raise a lot of funds by doing it.
Another workaround that firms would likely use is that they have 2
different streams of sending money back to the United States. One is by
repatriating profits, which this plan is designed to encourage more of.
(Of course, U.S. firms repatriating too much profits is the sort of
thing that anti-globalists complain about a lot, but never mind..) The
other is by having the affiliates pay the parents license and royalty
fees for technology generated by the parents’ R&D. So if I really need
the affiliate to be self-financing, I’d have them pay less license and
royalty fees instead to make up for the new tax incentives to repatriate
profits. Would this have an effect on corporate R&D, on productivity,
on American dynamism in general? Or would it just disappear in the
general fungibility of funds? This is the point at which I stop
pretending to understand it very well….