To: H-1B/L-1/offshoring e-newsletter 168

A couple of weeks ago the Wall Street Journal Numbers Guy, Carl Bialik,
did a column on the NFAP claim that each H-1B worker creates five new
jobs.  I reviewed that piece here in my e-newsletter; see
http://heather.cs.ucdavis.edu/Archive/WSJOnNFAPClaim.txt

One aspect Bialik brought up was patent applications by immigrant
inventors.  In response to Bialik's citing the patent study by Vivek
Wadhwa, I wrote that Vivek had withdrawn his study, and that Vivek "has
explicitly stated that he never found that immigrants are more
innovative than Americans."  Vivek agrees that the second half of my
comment was correct, but he has asked me to correct the first half.

Here is the background.  After Vivek released his patent study, I noted
that the apparently precipitous rise in immigrant patent applications in
the short time time of just eight years sounded highly suspect; things
just don't change so much in such a short time span.  Come on, immigrant
engineers didn't suddenly become much more creative than they used to
be, and their numbers in the industry didn't undergo a sudden sharp
increase.  So it just didn't make sense.  Moreover, Vivek's report
itself noted that 60% of the data was missing at the beginning of the
eight-year period of the study.  That really means that all bets are
off; no experienced academic would touch such data.  See
http://heather.cs.ucdavis.edu/Archive/WadhwaIII.txt

Then last July, Vivek wrote that he had found an apparent explanation.
Basically it was an accounting issue, stemming from the backlog of
applications for green cards.  He wrote that as a result many workers
had not yet changed from the "immigrant" to "American" category.  In my
review of that article, I expressed doubt that Vivek's theory was the
best explanation for the data anomaly (the sudden sharp increase in
patent applications), but that at least he does now agree with me that
the observed data pattern was not valid.  See
http://heather.cs.ucdavis.edu/Archive/WadhwaIV.txt

That's why I wrote in my review of Bialik's column that Vivek had
"withdrawn" his study.  That language was arguably too strong to begin
with, but Vivek feels that it was flatly incorrect.  He wrote (I'm
quoting his e-mail message with his permission):

#  What I have said is that I was baffled or confused or concerned about the
#  fact that foreign nationals contributing to WIPO PCT applications increased
#  337% over an 8 year period (1998-2006). And this was explained by the
#  increases in temporary workers without a corresponding increase in
#  green-card numbers. So they are coming in on temporary visas and may be
#  leaving. My view is that either we bring them in on the right visas or don't
#  bring them in. (I know that option 2 is preferred by many!).

This seems a bit different from what he wrote originally (see the above
URL), and one would think that an exodus of the foreign workers would
deflate the number of patent applications, counter to the sharp increase
that Vivek had found. 

Thus I still don't think the offered explanation is correct, but the
main point is Vivek did NOT withdraw his study, and still stands by it.
Vivek writes:

#  I had this triple checked by other academics and it was rock solid. We
#  examined every single record in several years of WIPO filings and double
#  checked the nationality information. This was not a random sample, but
#  an examination of the entire database for certain years.

Vivek also says:

#  You took Bill Kerr on, but didn't respond to his rebuttal. And you
#  didn't comment on Jenny Hunt's work.

Vivek is of course referring to the recent working paper by Harvard
professor Bill Kerr on immigrant patenting.  I had written a critical
review of that paper, which Bill Kerr responded to in an eight-page
letter to me.  I've said that I will post the letter (with his
permission) and my comments as soon as I get time to do it.  Unfortunately,
unlike Vivek and Bill, my "day job" is not to do analysis of the immigrant
tech labor force, and I just haven't had time to devote to this yet.  But
I definitely will do so, and will discuss the Hunt paper too.

In the meantime, I do want to emphasize that all three
researchers--Wadhwa, Kerr and Hunt--make no claims that immigrant
engineers are more creative or patent-prone than natives.  All three
have made specific disclaimers in that regard.  In the Hunt paper, for
example, the authors state in the abstract (emphasis added) "The 2003
National Survey of College Graduates shows that immigrants patent at
double the native rate, and that this is entirely accounted for by their
disproportionately holding degrees in science and engineering"
(http://www.econ-pol.unisi.it/~zanella/rgroup08/ImmigrationInnovation.pdf);
in other words, there are more immigrant techies, thus more immigrant
patents.

Norm