The cooperative patent classification

Anyone who has tried to use the United States Patent Classification (USPC) will know how notoriously difficult it is to understand, especially if your initial experience with patent classification is with the International Patent Classification (IPC) or the European Classification System (ECLA). The Cooperative Patent Classification (CPC) is a joint effort between the European Patent Office (EPO) and the United States Patent and Trademark Office (USPTO) to harmonise their existing classification systems. It has been in development for just over a year and is expected to be launched on 1 January 2013.


The CPC won’t be eliminating the IPC, at least not yet. Most countries have migrated from their own national classification systems to the IPC, with the US being the notable exception, but the CPC is a bilateral system with distant hopes of being introduced to other jurisdictions. I expect that the IPC and CPC will eventually merge. The CPC is another step in getting the US closer to the rest of the world, and soon it will make sense to have a single comprehensive patent classification system.

The benefits anticipated by the CPC for Europe and the US could easily be applied to the rest of the world’s patent offices: enhanced examination efficiency, improved access to documents from other offices, improved consistency of search results and increased work sharing through developments such as the various patent prosecution highways.

If you are familiar with the ECLA you will know that it is similar to the IPC but with many more subdivisions. The CPC will be similar to the ECLA with its large numbers of subdivisions but it will follow the IPC in the way its subdivisions are labelled.

In the ECLA the subdivisions are indicated with letters and numbers, for example, the IPC subclass A61K9/20 relating to pills, lozenges and tablets is broken down into a further 21 subdivisions in the ECLA such as A61K9/20H6F relating to pills containing polysaccharide excipients or A61K9/20K4 relating to layered tablets. Revisions of the IPC have tended to expand some of the groups, and in that expansion WIPO have indicated their subdivisions with extra digits which act like numbers after a decimal point. For example, in the IPC 4th edition, C07F9/65 relating to phosphorus compounds containing one or more N-containing rings is followed directly by C07F9/66 relating to arsenic compounds, whereas in the IPC 5th edition and beyond it is followed by C07F9/6503 relating to 5-membered rings with two nitrogen atoms, and another 30 more before we again get to C07F9/66.

This difference in labelling will require a complex conversion process involving the IPC, the ECLA and ECLA indexing codes, but that will be nothing compared to the shift those used to working with the USPC will experience. As far as I can tell the USPC is an upside down semi-hierarchical system, and I imagine those used to it, consider the IPC in the same way. Patent examiners in the US are required to classify by IPC as well as USPC, but this is usually done using the USPC-to-IPC Reverse Concordance tool, and it isn’t completely accurate due to the difference between the two systems. Many US patent applications have been classified into the ECLA by the EPO, and the USPTO has outsourced its reclassifications of the IPC to other offices such as the Korean Intellectual Property Office who are more fluent in the IPC, but at some stage they are going to have to classify documents into the CPC by themselves.

One thing that should help are the CPC definition documents being developed by the EPO classification experts. If you have ever seen the notes at the beginning of the IPC groups you will know they go into some detail so that classification of similar technologies is consistent. As good as the notes are I hope the CPC definition documents are even better. Some explicit examples would help.

From a searching viewpoint the CPC has its positives and negatives. The biggest negative, which is the only negative and really quite small, is that there will be an extra field to consider when developing a search strategy. I’m familiar with the IPC and ECLA so the leap isn’t great, and anything classified in the CPC will be classified in the same place in the IPC or ECLA so is likely to be captured by existing search strategies already rather than add substantial numbers of hits to the search results.

The positives include not having to try and classify using the USPC, and better access to foreign prior art. Eventual merging of the IPC and CPC will provide a much more detailed system, allowing for a more focussed search, which will become a necessity with the ever increasing number of patent applications being filed.

The CPC is aiming to incorporate the best classification practices of the EPO and the USPTO into a single system, and although it’s not directly relevant to searches on Australian applications, I hope they succeed.