State of Innovation

Patents and Innovation Economics

OECD Patent Quality Measurement – Confusing Patent Quality with Innovation

This study confuses patent quality with invention quality, Science and technology: falling patent quality hits innovation, says OECD.  The report makes sweeping statements about the decline in patent quality over the last two decades and suggests this is because of the rush to file patents on every minor improvement.[1] .  I can receive a high quality patent for a minor (non-ground breaking) invention.  The two questions are unconnected.  If the study has any meaning, it is about whether the world is creating revolutionary or disruptive inventions- not about the quality of the patents being issued.  If the report had clearly stated that they are concerned with a general lack of innovation, I would agree with this point.  After all, I wrote a book about this, The Decline and Fall of the American Entrepreneur: How Little Known Laws and Regulations. I show that the reasons why we are falling behind technologically are because we have weakened our patent laws and because of laws like Sarbanes Oxley.  Weakening our patent system makes it harder for inventors with revolutionary or disruptive technologies to raise capital and increases the risk to investors.  Sarbanes Oxley makes it almost impossible to go public in the United States, which increases the time necessary to obtain a return when investing in technology startups.  Startups are the source of most new revolutionary technologies according to the SBA.

Because this report is going to be trumpeted by anti-patent crusaders, I will address the patent quality issue.  The video on the report states that OECD has come up with a “new” way to measure patent quality, but neither the video or the text explains the methodology clearly.  By reviewing multiple articles it appears that their methodology quantifies patent citations and looks only at the top 1% of patents by the number of citations for that patent.  It appears that there are fewer citations for the top 1% of patents than the previous decade.  From this, they conclude that there is a 20%  overall drop in quality.  I assume this means there are 20% fewer citations for the top 1% of patents.

First of all, patent citations do not measure patent quality; but they may measure the importance of an invention.  Patents are cited in later patents for a variety of reasons.  I know many large companies purposely cite their own patents to inflate these numbers even when the earlier patents have little or nothing to do with the present patent application.  Second, by only looking at the top 1% of patents the study has arbitrarily defined its population.  Is there any justification for this population?  Can they show that this population is a good indicator of patent quality  or subsequent technological progress?  Did they look at a representative group of these patents’ file histories and claims?  Third, this could be an indicator that there is no  dominate technology driving the economy.  In the 1990s, the economy was driven by the internet and most new technology was revolving around this phenomena.  As a result, it is likely that you would have more patents citing each other.  Fourth, does this study take into account the global economic decline that started in 2000?

The broad conclusions of the study are not supported by its methodology.  How do they jump from their definition of the top 1% of patents to suggesting people are filing patents on too many minor inventions?  Here is an example of one of the statements from the report that seems totally unconnected to its conclusions:

Patents from inventors in the United States, Germany and Japan are the most highly cited, which suggests they are true innovations being used by many firms in their products to generate further innovations.

But while these countries produced about 70% of the top 1% of highly cited patents between 1996 and 2000, their share had fallen to 60% five years later.

There is no one metric that can probably capture patent quality, however I have looked at a number of metrics and they do not support the OECD conclusion.  I measure my metrics over a longer timeframe, the metrics have  broader measures, showing the trend relating to patent quality over the long term has increased.

Probably the most important macroeconomic metric for patents is the amount of research and development (R&D) spent per patent.  As the chart below shows, the amount of R&D spent per patent

has increased from around $1.2 Million per patent to around $4 Million per patent.  The chart shows the total amount of money spent on R&D by the U.S., according to the National Science Foundation, and the number of patents issued to U.S. inventors per year,according to the U.S. Patent and Trademark Office.  One potential limitation to the data is that the amount of money spent on R&D is probably representative of large corporations.  Small entrepreneurial start-ups are less likely to be accurately recorded in these surveys.  This may account for part of the downward trend between 1985 and 2000.

How does this compare to other research?  According to a paper by the Federal Research Bank of San Francisco[1], real industrial R&D has been growing at 3.7% per year between 1953 and 2000, while the number of patents per capita has been growing at 1.7% per year over the same period.[2] See the chart below.

The paper notes that the number of citations received by the average patent has increased over the last couple of decades.  From 1975-1999, the number of citations per patent has increased by around 3.3% per year according to the paper.  This suggests that the value of the average patent has increased and that the “size of the technological change codified in the average patent is increasing.”
Another patent quality metric is the GDP per patent ratio.  An increasing amount of GDP per patent should indicate that the quality of patents is increasing, since the number of patents for a given amount of economic activity is decreasing.  The chart below shows that this is indeed the case.

last 60 years.  Note:  the chart used GDP in chained 2005 dollars from the Department of Commerce.

As the U.S. transitioned from an industrial economy to an information economy we would expect that the number of patents per worker increased.  Indeed,  it seems likely that in an information society we can expect that the percentage of workers devoted to inventive activities to increase.  An information society is likely to have more of its GDP tied to inventive activities and patents.  As result, it would only be natural that the number of patents per worker would increase and the number of patents per dollar of GDP would increase.  There is no reason to suggest that the number or patents per dollar (inflation adjusted) spent on R&D would decrease in an information society, however that is what the data shows.  This is most likely a result of the U.S. patent system increasing the threshold for obtaining a patent and becoming less friendly to independent inventors and small businesses.  This would mean that the U.S. patent system is becoming more like the European and Japanese patent systems.  Neither of these patent systems are friendly to independent inventors or small businesses.  This may not seem relevant, but numerous studies have shown that independent inventors and small business are much more likely to develop and patent important or significant inventions.  For instance, the paper “An Analysis of Small Business Patents by Industry and Firm Size” by the SBA, states that “small firm patents tend to be more significant than large firm patents, outperforming them in a number of categories including growth, citation impact, and originality.”[1]

 

Conclusion

The amount of R&D per issued US patent has increased fourfold , or 3.7% per year over the last 50 years.  The number of citations per patent increased 3.3% per year over the 25 year period form 1975-1999.  These metrics suggest that the quality of patents issued by the USPTO (United States Patent and Trademark Office) have increased over the last 50 years, contrary to the results reported by other researchers.  In addition, the critics of US patent quality have selected to compare the US patent system to Europe’s and Japan’s, both of which have proven hostile to small entities.  Since small entities are key to the US’s technology innovation, it would be folly to compare our system to theirs.

Here is another post, Patent Quality Myth, on point.

 

Science and technology: falling patent quality hits innovation, says OECD.


[1] Anthony Brietzman and Diana Hicks, “An Analysis of Small Business Patents by Industry and Firm Size”,  Small Business Research Summary, No 335, November 2008.

 

 


[1] Note that if true the AIA will just make this trend worse, since the AIA effectively eliminates the one year grace period.  This means inventors will have to rush to file before they have had time to perfect their inventions.  In addition, the most innovative group of inventors are individuals and startups which are disadvantaged by the AIA and therefore will file fewer patents.

[1] Wilson, Daniel, “Are We Running Out of New Ideas” A Look at Patents and R&D”, FRBSF Economic Letter, Number 2003-26, September 12, 2003.

[2] Note that the R&D per  patent chart uses total R&D expenditures while the FRBSF chart uses Industrial R&D.

 

Advertisements

September 22, 2011 - Posted by | News, Patents | ,

No comments yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: