Last year, we wrote about the importance of the Startup Act 2.0 in helping to create more jobs via startups (the creators of tons of new jobs in our economy), and now there's a new version, the Startup Act 3.0. The Senate bill is truly "bipartisan" -- sponsored by Senators Moran, Warner, Coons and Blunt. The key focus, as in the past, has been on solving the issue in which we are driving some of the best skilled tech workers to other countries to build companies, rather than bringing them to (or keeping them in) the US. Our visa policy for skilled workers has been flawed for way too long, and this takes us a step in a saner direction. With the President and both parties of Congress seemingly (finally) open to the idea of comprehensive immigration reform, hopefully something as basic as better immigration policies for skilled workers is no longer even remotely controversial. In fact, for the most part, it seems like a very large number of folks in Congress support such reforms. The issue that has tied it up in the past has been whether or not this needs to be its own thing, or if it needs to be tied to more comprehensive immigration reform. And, really, I would imagine that's still the same stumbling block this time around, but with more appetite for that, hopefully the Startup Act 3.0 can be a part of all of that.

In terms of specifics, the bill would boost new visas for foreign-born entrepreneurs (these are people who create jobs, not take them from others) as well as for foreign graduates of US universities who have advanced degrees in science, technology, engineering and math. We train them, we might as well get the direct benefit of their work after they graduate. This is different than the issue of H-1B visas for skilled immigrants -- which is also an important program, but one that has been abused too often by companies. In this case, the visas are for the individual, not the company, and it focuses strongly on immigrants who create companies and jobs. As with previous startup visa efforts, I do have one minor quibble, which is that the entrepreneur visas will require you to have raised at least $100,000 in outside investment (or for an investment business, have invested that much). That's lower than we've seen in past proposals, but still seems to predicate the idea of doing a startup with raising outside capital. In this day and age, that's less and less necessary for many types of companies. Similarly, they need to have hired 2 outside full time (non-family) employees, and by the end of four years, increased employment to at least five non-family employees. The bill is useful for many foreign born entrepreneurs, though I still would have liked to have seen a test for entrepreneurs who don't want to raise money, but are building successful bootstrapped businesses.

The bill also has a few other provisions designed to help startups, including making permanent a 100% exemption on capital gains taxes for investments held for at least five years in "qualified small businesses" (i.e., a way to encourage long-term investing in small businesses) and creating an R&D tax credit (up to $250k) for startups less than five years old and with less than $5 million in annual revenue. Basically, small businesses get a tax credit for investing in research and development.

I'm also a bit concerned about section 8, which focuses on "accelerating commercialization of taxpayer-funded research." This one basically provides grants and funding to universities to do more commercialization. In theory, this sounds good, but I worry about it in practice. Most universities have, unfortunately, decided that "commercialization" means getting lots of patents and then selling licenses. In other words, the push to "commercialize" university research hasn't actually led to much commercialization -- but rather an awful lot of universities turning into pseudo-patent trolls. The plan leaves open some of the details to the National Advisory Council on Innovation and Entrepreneurship. Hopefully, they recognize that merely pumping up patenting and "tech transfer offices" is not really about commercialization. In fact, a better way to encourage commercialization might be to make federally funded (i.e., taxpayer-funded) university research public domain, rather than patented. That would actually encourage more usage in industry.

Either way, there's a lot of good stuff in the bill, and this time... they've even put out a snazzy infographic to go with the bill. How can it not pass after that?

This is unfortunate. Despite plenty of research showing that patents do not, in fact, lead to increased innovation (but rather increased patenting), many still assume that there's a direct linkage. Of course, it is true that many successful industries see high rates of patents, but there is evidence that patents tend to lag the actual innovation, rather than predate it. That is, once an area or industry is innovative and successful then everyone rushes in to get patents and try to extract their piece of the pie, often slowing down the pace of innovation.

So it's fairly disappointing that the Brookings Institution, which normally does pretty good work on these kinds of things has put out a study about patents and innovation, and appears to be confusing correlation and causation in saying that patents lead to innovation and even (more ridiculously) that areas that aren't doing enough patenting need to beef up their patents to increase innovation:

Metro areas that produce a lot of patents—and the inventiveness that that implies—are more likely to see above-average gains in population, productivity, jobs, and education, according to a report from the Brookings Institution, a nonprofit research and policy think tank. And the bottom fourth of metro areas, the ones that produce the fewest patents, could gain as much as $4,300 per worker over a decade if they amped up their patent production to match the top fourth.

“If we were able to get the roughly 250 metropolitan areas that do very little patenting up to the level of the 100 that do a great deal of patenting, we’d be richer in an extraordinary way,” says Jonathan Rothwell, a lead researcher on the study. “It would make really a huge difference to economic development.”

Since the report focuses on successful metro areas, it seems that there are many, many other factors that may have resulted in the successful economic situations in those areas, and those other factors may also have led to the increase in patenting. Assuming a causal relationship and (worse) suggesting that all other regions need to do is up their patenting, is a dangerously ill-informed suggestion. While the report claims to account for "reverse" causation, it appears to make little to no effort to really account for the many, many variables that are easy to observe in every day life that lead to a correlation between patents and economic output.

I'd been working on a response to some of the many methodological problems I spotted in the report, and it was growing ever longer and longer... and then I saw that Eli Dourado did a much better and more concise job of it in explaining why the report is bogus. Dourado points to two possible explanations for the correlation, neither of which are accounted for by the paper:

These conclusions are unwarranted given the model and findings expressed in the paper. To see that this is the case, assume temporarily that patents do nothing to incentivize real innovation, and that they merely transfer wealth from consumers at large to the patent holder through firm profits. If this were the case, then we would find that measured output per worker was higher in metropolitan areas with more patents—exactly what the authors found!—because they are gaining profits at the expense of consumers in metropolitan areas with fewer patents. In other words, the authors could be laboring under a fallacy of composition. Just because patents enrich the MSAs that generate them doesn’t mean that they are a source of prosperity for the nation as a whole or that they increase social welfare.

Alternatively, assume temporarily that patents do nothing to incentivize real innovation, but that firms that produce valuable innovations must defensively patent them to avoid being taken to court for using their own inventions. If this were the case, then patents would correlate with real innovation, and therefore with output per worker, but they would not cause an increase in productivity. In addition, at least some of the measured increase in output would come from an influx of highly-paid intellectual property attorneys, which by assumption does not represent real added productivity. Note that the top-patenting MSA in the study is Silicon Valley, the part of the country where people are most concerned about defensive patenting. But the word “defensive” does not appear even one time in the report, the appendix, or the working paper.

The authors have done nothing to identify the effect of patents on productivity, which is to say, nothing to rule out either of the possible assumptions above. They are simply relying on the assumption that more patents means more innovation.

This is a major major flaw in the paper. It seems to assume that a whole bunch of things that simply aren't seen by folks who actually work in the industry, and makes little to no attempt to account for those other variables. In the comments, the lead researcher on the paper, Jonathan Rothwell, tries to defend the paper, by saying (in part) that they're just using patents as a proxy on inventiveness, and the paper should not be seen as supporting patents or the patent system itself:

Right up front, I think it is important to keep in mind that our study aimed to examine the effects of invention rather than the effects of patents themselves. Hundreds of economic papers have been written that use patents as a proxy measure of invention (based on detailed firm and industry level analysis), so I think that is fairly uncontroversial.

Just because lots of folks do it, it doesn't mean it's right. But the bigger issue is that while he claims that the paper is not an endorsement of patents specifically or increasingly patenting activity, that's not what he's telling the press. Just look at the quote we have above from what he told the National Journal. He specifically is saying that we should boost patenting in other metropolitan areas, suggesting that it would make "a huge difference to economic development." In other words, contrary to his claims in the blog comments, when talking to the press, he's pretty clear that he believes there's a direct causal relationship between patents and economic development. Furthermore, if he really believed that, he should have disclaimed, publicly, the title of that National Journal article, which explicitly says that patents (not "inventiveness") "matters to economic and job growth."

Also... if the report really is about "inventiveness" and not "patents," perhaps the paper should not have been called "Patenting Prosperity." Just saying.

Brookings, of course, is quite well-established and respected, and you can bet that pro-patent-system folks will be using this report to claim that "more patents are better" and that any reforms that are designed to push back on bad patents or to start limiting the number of patents we issue, would be a bad thing. Even though the report does, in fact, contain some arguments in favor of limiting certain types of patents and patent system abuse, those nuances will undoubtedly be lost. This report is going to get cited repeatedly as "evidence" that we need more patents and stronger patents, despite the fact that the actual evidence says no such thing.