Wednesday, September 24, 2014

Do Patents Really Work?

The stated purpose of granting patents is to provide an incentive for inventors to innovate and to share their innovations with the public.  And as with all public policy it is fair to ask the question:  Does patent policy actually influence the behavior of inventors?  It turns out this is a tough question to answer.  While the anecdotal evidence suggests that patents are doing their job, a truly rigorous study of the question is hindered by two problems.

The first problem is that you really can’t measure behavior statistically.  You can only measure changes in behavior and changes in behavior have to be driven by changes in policy.  And since US patent law and policy has been pretty stable since the 1950’s there are only a limited number of changes available for study.

The second problem is that once you have a policy change one still has to be able to separate the behavior change caused by the policy from other societal factors that may drive changes in behavior.  Patent policy isn’t the only factor that drives inventive activity.  The number of patents granted will be influenced by economic activity, education and income levels as well as host of other potential factors. 

In 2010, David Abrams attempted to measure if patent policy can impact innovator behavior.  And in Did TRIPS Spur Innovation? An Empirical Analysis of Patent Duration and Incentives to Innovate he appears to have found an answer (yes).

For the change to the system, Abrams chose the 1995 implementation of the TRIPS agreement which was part of the GATT approval. Specifically, in the TRIPS agreement the US agreed to change the patent term from 17 years after the patent issue date to 20 years after the patent application date. 

Once Abrams chose the change, his next problem was to isolate the change from other societal factors.  What Abrams did was to use a “difference in differences” methodology.  Abrams described this methodology using an example.  Assume that bankruptcy laws were relaxed significantly for people who earn less than $150,000 per year but not for people who earn over that amount.  Then if, after passage, there is a 10% increase in the rates of bankruptcy the initial conclusion might be that the law increased bankruptcies. What Abrams points out is that there is a way to test this conclusion:  compare bankruptcy rates for people who earn less than $150,000 with those who earn more than $150,000.  His argument is that if the bankruptcy rates between the two groups both increased by 10%, then the law probably had no impact.  Conversely, if there is a significant difference in the increases between the two groups, then the law probably had a significant impact.

Abrams took advantage of the fact that the TRIPS increase in the patent term would be offset by the processing time the USPTO took to approve the patent.  Because these processing times vary by patent classification code, Abrams reasoned that if inventors were rationally influenced by the change in term, he would see more applications in patent classes that had shorter application processing times compared to those with longer processing times.  And, after analyzing over 800,000 patents, this is exactly what he found.  

It doesn’t look like this increase in patent quantity came from a reduction in quality.  Abrams compared forward citations, (a typical quality measure for patents) for patent applications made before and after the term change and found no significant difference.  Unfortunately, since Abrams was measuring differences instead of absolutes there is no way to quantify an absolute increase in applications from this study. 

To be fair, the value of an empirical study with this result is limited.  What makes this study worth noting is that by demonstrating that inventors and innovation can be influenced by patent policy, Abrams created a methodology that allows future researchers to separate the impact of patent policy from other societal factors.  What they will discover only time can tell.

Below is a summary of selected patents that have been recently issued in textile related classification codes:

Sun protective neckwear garment:  An ultraviolet protective garment to protect the user's neck, chest, upper back, chin, and portions of the shoulders. The neckwear garment is a single body tubular unit, with a ship decanter shape.  The neckwear garment is manufactured from flexible, elasticized, fabric with ultraviolet blocking capability. Patent:  8689362.  Inventor:  Lavin.  Not Assigned

Glove and method for producing the same:   A glove having excellent moisture permeability and abrasion resistance of a back portion, and does not trap heat or retain moisture.   A glove has a coating layer of rubber containing air bubbles formed on the back portion of the glove.  The moisture permeability of the coating layer ranges from 1000 to 9000 g/m2·24 hrs. Patent:  8689363.  Inventors:  Kishihara and Ii.  Assignee:  Showa Glove Co.

Belt adjustment system:  A heavy duty belt adjustment system that permit a continuum or larger selection of belt loop sizes..  Patent:  8689364.  Inventors:  Rowland and Hunsucker.  Not Assigned.

Fabric having ultraviolet radiation protection:  A method for treating a fabric for protection from ultraviolet radiation.  The method comprises washing the fabric in a washing machine with a suspension of zinc oxide particles treated with an acid polymer.  This binds the zinc oxide to the fabric.  Patent:  8690964.  Inventors:  Kramer,  Assignee:  The Sweet Living Group, LLC

Paper finishing process:  A process for applying a finishing paper material on a leather or woven support material is described, comprising the steps of: first buttering of glue on a surface of the support material, positioning the finishing material on said support material, ironing, perching, second buttering of glue, and a finishing step.  Patent:  8690965.  Inventor:  Magrin and Pellizzari.  Assignee:  G & G S.R.L. (Societa Unipersonale)

Tobacco plant derived dye and process of making the same:  A dye and process to make a dye produced from tobacco plant materials In a preferred embodiment, the tobacco plants are organically grown and the stems and leaves of the plants are utilized to produce the dye. The process percolates a hot enzyme solution onto ground plant materials to produce tobacco plant fiber and the dye. The dye can be produced in many colors, does not require a mordant to bind and does not produce waste..  Patent:  8690966.  Inventor:  Devall.  Assignee:  PBO, Inc.

High tear strength flame resistant cotton fabric:  An affordable flame resistant cotton fabric with better tear strength. The 100% cotton yarn is produced using a combination of compact spinning technology and gassing and/or singeing process performed in tandem.  The yarn is turned into fabric through normal processes.  A flame retardant is applied with appropriate finishing chemicals.   Patent:  8689413.  Inventor:  Ramaswami.  Not Assigned.

Jim Carson is a principal of RB Consulting, Inc. and a registered patent agent.  He has over 30 years of experience across multiple industries including the biotechnology, textile, computer, telecommunications, and energy sectors.  RB Consulting, Inc. specializes in providing management, prototyping, and regulatory services to small and start-up businesses.  He can be reached via email at or by phone at (803) 792-2183.

No comments:

Post a Comment