The United States Patent & Trademark Office collates and makes available online statistics relating to patent office activity. The data provided by the patent office could be thought of as a mine of information. A few companies make a business of collecting USPTO data and subjecting it to analysis.

One of the more interesting things to be found is the % fraction of patents granted to foreign entities. As of 2007, the fraction of allowances to foreign entities is 49 %. The above graph shows the tend over time. The gap in the curve is due to the absence of data for 1975-1976 in the published data set.

The fraction of foreign patent allowances has remained approximately constant since ca 1985.  There was a dip in the 1990’s that may correspond to some sort of pullback in R&D activity. This drop off in issued patents lags by several years due to pendency.

The above graph uses data published by the USPTO. Here we see the accumulated patent allowances to various nations over the period from 1963 through 2007. The data set is limited to Organic Compound classes 532 through 570 under the US classification system. For brevity, only the top 8 foreign applicants are shown in comparison to the US.

Very obviously Germany and Japan have the leading foreign awardees of US patents in this segment of R&D. I have not looked at how the reciprocal situation compares under PCT filings abroad by US applicants. 

What is of greatest interest is seen in the top graph: 50 % of the patent real estate being staked out at the USPTO is going to foreign interests.

A unique feature of chemical patents is the Markush claim. Markush claims allow the claiming of a  potentially large huge number of analogs defined by compact symbolism and covering vast swaths of the periodic table. 

It is thus possible for a professor in Osaka to own the composition of matter of a Markush set of hundreds or thousands compounds that would then bar a company in New Jersey from making even an obscure member. Under the PCT, the same is true in the other direction.

The reach of property rights in the world of invention has become so extensive, and the data provided by the various patent agencies and abstracting services is so inadequate, that the act of performing a due diligence search is nearly comical. In truth, you look for low to middle hanging fruit and hope that an obscure sentence somewhere doesn’t blow a hole below your waterline one day.

Chemical patents are in dire need of reform in terms of the nature of the disclosure. Patent offices must find a way to facilitate the extraction of crucial information so the public has a fair chance of understanding what is off limits.

We need a more lucid recitation of claimed compositions and better use of language in the detailing of processes. Patents should be written with abstracting in mind. It should be made possible to extract processes and compositions into a form that can be accumulated in databases for rapid review. This has to begin in the drafting phase of the application.

[For some great feedback, check out the comments- Th’ Gaussling]