Prof. Tom Brennan on the tax holiday
Insightful, valuable new research by NU Law prof Tom Brennan (reported here).
Here is how Tom describes his project:
I think there are at least two important components of this work. The first
is the finding that firms largely followed stated legal restrictions on how to
spend money, even though fungibility of money and lack of tracing requirements
would have allowed them to get around the restrictions and still remain within
the technical confines of the law. They thus followed the spirit of the
law to a greater degree than bare technical compliance would have
required. This can makes sense, though, because the firms are long-term
repeat players in a game with the government, and they don’t want to be seen as
egregiously violating the rules in one round because this might worsen their
treatment in a future round.
The second is the fact that my work proves that prior research, published in
the Journal of Finance and cited in a Senate report, has been wrongly
interpreted. The prior research says that 60 to 92 cents per dollar
repatriated was spent on shareholder payouts (dividends and share repurchases)
in 2005. This simply isn’t true, and it is easy to prove that it is
wrong. If you just add up all the spending on shareholder payouts in
2005, you get a fraction that is less than 60 cents on the dollar of
repatriated funds. It is a rare thing to be able to prove that an
interpretation of statistical results is definitely wrong, but this is such a case.
I want to be sure to be fair to the original Journal of Finance article.
Their findings can still be correct, but the interpretation has to be clearly
cabined as something that speaks to the marginal behavior of a typical firm of
a certain sort, and not a statement about overall spending of dollars.
This is a distinction that they did not focus on, and as a result
interpretations of the results effectively conflated results about a typical
firm of a certain sort with results about typical dollars. In fact, most
all of the dollars were brought back by a very few firms, and so understanding
how a typical dollar was spent requires a careful understanding of these few
Two further articles citing Brennan’s work are included here for your reference: