It’s no secret that when David Kappos took the reins of the USPTO in mid-2009, a major focus of his administration would be on reducing the backlog of outstanding patent applications significantly. Now that his departure as head of the USPTO is two and a half months cold, did he succeed?
This pre-Kappos article from IPWatchdog bemoaned the ever-rising time it took to hustle a patent application through the examination procedure. From the 1980s through the mid-1990s, the average utility patent took around 18 months to pass through the office. When Kappos was sworn in August 2009, this was notched up to just above 32 months, with over 800,000 applications awaiting a first office action from an examiner.
Seemingly overnight, the USPTO switched gears into examination overdrive mode. This PatentlyO post details how within two years, the backlog had dropped to just above 650,000. A recent paper, after some fancy statistical footwork, pegs the 2012 patent application acceptance rate at almost 90% after accounting for refilled continuing applications.
The USPTO offers a wealth of raw historical data, making amateur statistical analysis a breeze. I decided to run a quick model to see for myself how the allowance rate has changed. Using yearly filing and allowance numbers and assuming an (optimistic) pendency of two years, I arrived at the below graph.
USPTO Utility Patent filing numbers and allowance rates, 1988-2010.
All this analysis and hoopla about allowance rates begs the question: is this really the right way to evaluate success? Does it all really all come down to reducing the backlog and shuffling patents through the system? Shouldn’t overall quality of the allowed patents be the go-to metric?
Something so nebulous as how much a granted patent “promote[s] the progress of science and useful arts” is obviously a non-starter, short of an in-depth economic analysis of gargantuan proportions. What about the rate at which patents are ruled invalid by the courts?
Now we’re on to something. If we define the quality of a patent (or at least, the quality of the USPTO’s examination procedures) based on the percentage of patents invalidated or upheld in court over time, this would be a true measure of the USPTO’s efficiency – how many false positives (patents that should be granted, but aren’t) and false negatives (patents that shouldn’t be granted, but are) it issues.
This would obviously be quite the project. Do I hear any fresh law review members across the country looking for a note topic?
The post Post-Kappos – Rethinking USPTO Efficiency appeared first on Beyond Clause 8.
