The key point in analyzing patents is the fact that years of exploring statistical evidence by economists hasn't found any evidence they work to encourage innovation. Kevin Smith of Duke - who is very interested in copyright - has been pouring over economic data and stumbled over the same fact. Facts are stubborn things.

(hattip: Ruth Lewis)

David:

Kevin did not use the word "innovation" in his article. His article was related to economic growth. There are other articles, but the one I have yet to see "debunked" is:

Albert G.Z. Hu and I.P.L. Png, in their extremely well researched paper "Patent Rights and Economic Growth: Cross-Country Evidence," found a positive mathematical relationship between patents and economic growth. From their conclusion:

Using an ISIC 3-digit industry level database that spans 54 manufacturing industries in 72 countries between 1981-2000, we found evidence that stronger property rights were associated with faster industrial growth measured by value added. The impact of stronger patent righs was both statistically and economically significant in three of the four periods we analyzed: 1981-85, 1991-95, and 1996-2000, and had become stronger in the 1990s compared to the in the 1980s.

Our analysis also showed that the stronger patent rights promoted industrial growth through technical progress in the 1981-85 and 1996-2000 periods and through more rapid factor accumulation in the 1991-95 period.

Be cautious with recent working papers. As I wrote Ivan and they are investigating - there is a question about how foreign direct investment figures into this. We know that patents attract foreign direct investment. We know that foreign direct investment leads to growth. So it may be that patents do not promote growth through increasing innovation - many many studies find they do not - but that they promote growth by rearranging the flow of investment. To the extent this is a zero sum game between countries it doesn't really tell us whether patents are a good or bad idea.

There is also a question of causality: did strong patents promote growth? Or growth promote patents.

The raw correlations in the data are well known. Careful studies that try to penetrate the surface have found nothing underneath.

Terminology may be a problem. Considering the definition that some people have of innovation, patents may or may not have any direct relationship to innovation at all. I suppose one could make the same argument regarding scientific discovery:

Scientific discovery does not have a direct correlation to innovation. Ergo, scientific discovery is unimportant to innovation.

The problem is one of timing. A scientific discovery may not show up in an invention for decades, and perhaps not centuries. The basic science for lasers and televisions was developed in the latter half of the 19th century, but it took us decades to invent the materials that actually made those scientific discoveries implementable.

There was also a hug time lag between invention and innovation, as some define it. Did patents slow the dissemination of television into society? If we are separating the reason why it took so long for television to spread throughout the United States and the world, patents might be credited with a fraction of a percent of that reason. It is likely that the effect of patents is unmeasurable. On the other hand, innovation was slowed by lack of inventions because there were a lot of things we did not yet know how to do, and by lack of infrastructure, not to mention the relative unavailability of television sets.

Thus, there may be a huge lag between invention and implementation for reasons that have little or nothing to do with the patent and everything to do with infrastructure and associated technologies that need developed to exploit a new invention.

It took so long to get all the associated technology related to cell phones invented and people interested in using that technology that by the time the first cell phone system was actually ready to go, Motorola's patents were close to expiration. All this in spite of the encouragement Motorola gave to a host of companies in developing related technologies. Of course, when delving into a completely new invention, such as cell phones, it can take a long time to bootstrap the activity into being cost-effective and acceptance is never certain, so only strategic thinkers are interested in supporting the inventions that ultimately led to "innovations." The lack of patents in areas such as this only discourage investors because it increases risk and reduces the chance for an ROI. I wonder how many decades it would have taken for cell phones to be implemented had patents not existed? Of course, the same might be said for many other technologies developed in the last 100 years.

If patents in fact speed up innovation we'd expect countries and time periods with strong patents to have over a period of decades higher innovation rates than countries with weaker patents. The evidence shows this is not the case. It is certainly hard to prove that something would have happened slower without a patent. Of course there are lots of examples of things that happened slower because of patents - steam technology, automobile technology, the airplane...and on and on and on.

You are very focused on the idea of "wouldn't do that if not for the incentive of getting a patent." This is offset by the "can't do that because of someone else's patent." The data indicate a rough balance between the two, with a net effect of the patent system of about zero effect on increasing (or speeding if you prefer) innovation.

That isn't an argument for a patent system: patent systems do a lot of harm. The only reason to accept that harm is to get increased innovation. Unfortunately that doesn't work.

I disagree that having increased innovation is the only reason to have patents. As Thomas Jefferson noted, patents seemed to beget more invention - and good ones at that.

There is separate evidence in modern times that implementation of protection for plant yielded substantial numbers of new plants in both U.S. cotton industry and in China. In China the difference was orders of magnitude. The lessons we learned from this is that changing from no protection to patent protection can give you more of the thing protected - in this case plants. The protection was extended because there was a perceived social need for more plant varieties, and producers were uninterested in developing the varieties because they knew they would be copied immediately on production and the producers did not think they would be able to recoup cost. Sounds familiar.

As for your example of airplanes, steam technology and automobile technology, I note that:

(1) These are three anecdotes. There are millions of patents, and extending three anecdotes is not evidence that the system failed or is failing.

(2) Steam technology revolved around Newcomen's configuration for more than 50 years. James Watt figured out a better way. If his technique was that obvious, surely someone else would have figured it out a little faster than 50 years. What if Watt's configuration had not been invented for another 30 years? Watt's invention started the clock ticking until all could use the invention freely. Before the invention, there was no clock.

(3) James Watt and the Wright Brother are routinely used as examples of what not to do as an inventor. "Invention & Technology" magazine has had excellent articles on how Watt and the Wright Brothers managed to act against their own best interests by not optimizing exploitation of their patents. The same goes with any profession. Having an advantage never means it will be used wisely. I see that every day on the road. Just this morning a woman hit a motorcycle at an intersection. If we are to take away people's power just because they abuse it, then perhaps we should all be locked away lock the people in "The Matrix."

As for the patent system, there is substantial evidence that it has been and continues to be beneficial. Does it also or can it also do harm? Probably. So can virtually everything invented by man. Is the potential for doing harm the reason to ban something, or should we learn to use our tools wisely? Why not ban explosives because do a lot of harm? What you have failed to do, as have others, is to create a T-column with the benefits on one side and the disadvantages on the other to see how the two sides sum. Until you do that, I see your comments as a biased diatribe without objective support - though I remain optimistic that someone will approach the question of the value of patent objectively and provide a true summing of the columns.

David:

Why do you believe that countries with "stronger" patents have lower innovation rates than countries with weaker systems? According to the chamber of commerce, the countries that have the "strongest" patent systems also tend to be the most innovative.

According to the chamber of commerce, the countries that have the "strongest" patent systems also tend to be the most innovative.

Correlation doesn't prove causation. The U.S. has a stronger patent system than Zimbabwe, for example, but I doubt that the former's better record of innovation is caused by its stronger patent regime.

I should have mentioned, following on David's point about the negative impact of patents on innovation ("can't do that because of someone else's patents"), that property rights and market-based institutions have no such negative effects on innovation. And property rights and free market institutions are the dynamic results of spontaneously evolving order, unlike patents, which are government-granted monopolies.

Bill:

Correct, correlation does not prove causation. However, the report generally shows a correlation between patent systems and levels of innovations, with only a few exceptions, and none of those exceptions are significant economies.

As for your other comment regarding patents, I counter with "you can do that because of an invention that was created, patented, and licensed or the patent has expired." Property rights and market-based institutions do have a negative effect on innovation. I will name two significant hinderances to innovation throughout most of the 1990's: Lack of skilled manpower and lack of capital.

As for patents being a "government-granted monopoly," we must never forget that the full title is a "government-granted monopoly authorized by the people of the United States." As I have pointed out before, intellectual property rights, which were evolved to encourage a societally beneficial level of invention, and which continue to evolve, are a people-empowered property right. When the people are convinced that patents are insufficiently valuable to grant, they will have intellectual property rights revoked.

I'm not sure you want to use the Chamber of Commerce as a neutral source of analysis. Check our book for a review of many studies by many different economists, or Josh Lerner's website for his recent work.

Property rights and market-based institutions do have a negative effect on innovation. I will name two significant hinderances to innovation throughout most of the 1990's: Lack of skilled manpower and lack of capital.

A lack of capital and skilled labor during the 1990s--the doc.com era? No way, Jose. Every vc and her mother were throwing $$ at tech start ups, and those were the ones that didn't float shares that erupted in price the first day of trading thanks to a gullible and punch drunk public market. As a tech blogger wrote a while back, during that decade you could have dressed up your dog in a tie and gotten him a job in a tech start up. Complete with options if not a life time supply of Kibbles and Bits.

Regarding your second point, the government is not the "We the People," as school children are taught in the government indoctrination camps known as public "schools." See Lysander Spooner ("Vices Are not Crimes"), and especially the first paragraph of Murray N. Rothbard's essay "The Anatomy of the State". Reading this essay instantly converted me from a dissatisfied and searching standard-issue lib to a full bore anarcho-capitalist.

Government is s lie and a delusion. A criminal gang, as Rothbard put it in For a New Liberty.

David:

I am not sure I want to use the studies in your book as a neutral source of analysis either. They contain biases of their own. I am unfamiliar with Josh Lerner's work, but may check him out at some point.

Bill:

Foolish people may have been throwing money at the dot com's, but in the real world (those of us who make actual products you can touch), engineering talent was extremely difficult to find in the latter half of the 1990's and then again in the period from about 2003 to 2007. Money has been a little easier to obtain, but has been tough from time to time. On the other hand, making stuff out of metal has never been as sexy as the internet.

I disagree with your statement regarding the government. The American people have proven that when they are actually interested in making change, they can. The problem is not that they are unable to make change, the problem is that they are not interested in change. However, if you can galvanize the people, you can do anything. Reagan did it more than once, to Congress's chagrin. Others have done it as well. While our democracy has issues, it also has opportunities, if the right person is there to establish a worthy goal and gain the support of Joe Average.

As for anarchy, you will never get more than a fraction of the population to support an anarchist type of government.

The studies in our book aren't our studies. They are all the empirical studies by economists that we could find. (The P'ng study came out after the book was published.) Unlike the chamber of commerce these are scientific studies. They are not carried out by advocacy groups, nor do the people that conducted the studies have any financial interest in patents. Indeed, most of them accept the conventional economic view that patents are essential for innovation, and they are extremely puzzled they can't find evidence that patents increase innovation. Nor are they monolithic in their conclusions.

David:

I struggle with the whole link to innovation thing. Invention and innovation are linked, but not necessarily in a direct way. As has been noted before and by others, there is no direct link between basic reasearch and invention. Similarly, there is no direct link between invention and marketable products. However, there can be a direct link.

For example, the Hall Effect was discovered in 1879. It took about a hundred years for materials science to be able to apply the Hall Effect (invention), and today we have thousands of products that use Hall Effect devices (innovation).

Clessie Cummins developed the engine compression brake after more than a couple of decades and thousands of dollars later, he had an invention. However, as happens with many inventors, his goal was a marketable product. After assuring that his invention would not be copied by filing of a patent application, he approached companies to see if they were interested in producing and selling his highly inventive and innovative production (innovation). Hundreds of thousands of brakes later, we know that Clessie's patented invention was a smart move for society.

So, while there is not necessarily a direct link between any of the steps that take us from science to innovation, there certainly can be. However, I would not look for one. Someone might as well try to quantify the relationship between basic science and innovation. Based on the difficulty of linking them, one might argue we should perform no more basic research, which would be short-sighted stupidity. Invention merely sets up a basis for innovation, it does not DRIVE innovation, which is ultimately a pull rather than a push, even when marketing efforts are highly creative.