Second-Class: A review of the trademark classification system

New Zealand, with around 150 other countries, applies the NICE classification system as part of its trademark regime.

Trademark applicants specify what class/es their trademark is to be registered for based on the goods and/or services it is intended to be used on. The trademark classification system has certain administrative benefits. It makes it easier to search for identical or similar marks already registered in respect of same or similar goods or services. It allows for IP offices to charge per class applied for, so the cost is roughly proportionate to the monopoly being claimed. And the NICE classifications have been adopted by most jurisdictions which simplifies the process for applications across multiple countries. However, there is a real question of whether these benefits still justify a classification system or whether in the present day, its failings outweigh its strengths.

The NICE classification system is showing its age. Agreed in 1957 but devised in 1935, the world has experienced significant change in the 87 years since its genesis. Entire new sectors exist today as the result of innovations that could not have been predicted in the pre-World War 2 era. Nuclear power, space travel, radar, jet engines, digital computers, mobiles and most significantly the internet have completely revolutionised industries as well as creating their own. To illustrate this point, if California – home to Silicon Valley – were an independent country, it would have the 5th largest economy in the world.[1]

The original justification that a classification system makes searching easier is not as relevant in the modern context. In the days of manual searching through catalogue cards, it made sense for trademarks to be grouped together in classes because it was easier to determine whether a similar mark was already registered for a same or similar product. However, this efficiency of process is negligible in the context of modern computer database searching. Nowadays, patent officers or any member of the wider public can simply type the features of a trademark into the trademark database and access relevant results instantaneously. In contrast to patent searching – where IPC classes are generally speaking essential for sifting through the sheer volume of information stored on patent databases, the nature of trade mark searching – to a large extent an “at a glance” exercise – combined with the ability of modern computer databases, means the Nice Classification is, if not redundant, then at least nowhere near as important as it used to be. That is even before considering options to improve search algorithms, as discussed below. The classification may even be irrelevant information when searching – the proviso “same or similar goods or services” means a substantially identical mark may not infringe another mark in the same class because the products are sufficiently distinct, or on the flipside, may infringe another mark in a different class because the products are related enough for there to be a risk of confusion. 

The classifications themselves are arbitrary and material-specific, so do not correlate well to industries nor to products of more complex manufacture. Some products do not fall neatly into any category. Others are included in multiple. For example, knives could be in class 7, 8, 10 or 16 depending on the intended use. Some class groupings are illogical – like beer being in a different class to all other alcoholic beverages. Applicants whose product does not fit neatly into a class are forced to apply under multiple which imposes extra costs. The additional cost per class therefore becomes a reflection, not of the extent of the monopoly sought, but merely the incapacity of the classification system to accommodate the desired product. Another issue is that the past few decades of technological innovation have made some classes hugely broad and others nearly redundant.

Class 9 is a clear illustration of the flaws of the classification system. The class covers a wide and disparate range of goods – from computer software to swim goggles, flash lights to fire engines, mouth guards to magnets, and spectacles to dog whistles. A trademark on dog whistles that is near-identical to one on computer software is highly likely to be granted, despite both belonging to Class 9. The class no longer contains “similar goods” and this is the fundamental premise of the classification system. Not only does class 9 have the widest diversity, it also has the largest volume of applications, by a significant margin. There were over 582 000 trademark applications filed in class 9 at the USTPO between 2009-2019. In comparison, Class 23 had only 3 600. In fact, there were more Class 9 applications filed in the first three months of 2019 than have ever been filed in Class 23. This volume and variety means the categorisation of Class 9 no longer conveys any meaningful information; its function is near obsolete.

Not to suggest the Nice Classification has remained “frozen” since its creation. It is periodically updated to include new subject matter. As an example, “humanoid robots with artificial intelligence for use in scientific research” is now included in class 9. As another example, from 2002 the 8th edition of the Nice Classification created new classes 43, 44, and 45 from reallocating some of the services in class 42. This was further refined under the 9th edition when, from 2007, legal services moved from class 42 to class 45. But these updates are piecemeal at best, and do not address the problems discussed above. AI technology was just lumped into an already radically-overpopulated class – class 9 – further exacerbating the problems in that regard. Simply “grafting on” new entries or subcategories is no meaningful solution to the problem. A smarter approach – especially when it comes to the sagging class 9 – is required, to ensure the Nice Classification actually delivers on its promised benefits.

Looking Forward

The golden question is then, are we stuck with these flaws or is there a way to reshape or recreate our trademark system?

The obvious answer is to modernise the classification system. Redevise the classes to better reflect modern society. Assign computer technology its own class. Put beer with wine. Make the classifications more industry-specific and accommodating of future innovation. However, there are two problems with this approach. Firstly, back in 1935 they would have thought they were devising the best classification system. The point being, the future cannot be predicted. Innovation continues to happen at an ever-increasing rate. Re-classification simply prolongs the problem as in another 40 years, industries may be reshaped yet again. The second issue is that it leaves all trademarks granted under the NICE system in a form of limbo. It would be an immense job to re-classify them. Therefore the workload of trademark applicants and examiners essentially doubles – they would have to search both the old NICE database, and the new reformed one. This already occurs to a certain extent when the NICE classifications are periodically updated. However, complete reconstruction of the classification system is on another scale entirely and would pose a significant burden on searchers.

The choice appears to be maintain or abolish. If the NICE system was retained, some tweaks would be desirable such as subdividing Class 9. But the core issues still remain. Abolition of the classification system entirely is radical but arguably the best way forward. Guidance can be taken from Canada which operated this way until 2019, when it conformed to the NICE system as a requisite of joining the Madrid Protocol. The Canadian trademark regime simply required goods or services descriptions to be worded in “ordinary commercial terms” without imposing any external classification system. The result was that Canadian-registered trademarks generally had a broader scope and cheaper cost than other countries. But as previously emphasised, this concept of ‘cost proportionate to claim’ does not really reflect reality. Adopting the Canadian approach of describing the product in “ordinary commercial terms” would mitigate the concern that without classifications, trademark owners would get all-encompassing rights that would block other trade. 

Those averse to abolishing the classification system have two other key objections. Firstly, that trademark searches, even electronic, would be too difficult without classes to refine the scope. Secondly, that products can be described in different ways, particularly in different countries and this issue will be exacerbated as the market becomes increasingly global. Galoshes, wellingtons or gumboots anyone? There is one answer to both these concerns. Investing time and resources - that would be otherwise spent working on the current system – into developing more sophisticated software to compare goods, services and trademarks. Google and social media are founded on the ability of a search algorithm to give the user what they are looking for. Development of better search algorithms that can cross-reference terms for the same product; analyse an image instead of just words; and sort and report on findings – this is the future of the trademark system. Since 1935, information technology has advanced significantly. The NICE classification was designed to address an issue that, 80 years on, we are able to resolve in an entirely different, and far more effective, manner. Instead of letting technological developments frustrate our trademark classification scheme, it is time to harness them to create a more effective system that will stand the test of time.

After canvassing the issues and solutions, it is clear that the trademark classification system no longer provides an important service to those involved in the trademark process. The essential need is for a trader to know that they can use their new trademark without infringing on anyone else’s. There are more effective and long-lasting solutions than a classification system that dates faster than the iphone. Technological innovation should be the trademark system’s asset, not its downfall. Powerful search engines and a practical approach that takes guidance from Canada, in replacement of the classification scheme, is the way of the future for the trademark system.

[1] Tim Lince “The dominance of Class 9: is it time for a radical overhaul of the Nice Classification system?” (04 December 2019) World Trademark Review <worldtrademarkreview.com>

Author: Hannah Houghton, Law Student

//piperpat.com/news/article/second-class-a-review-of-the-trademark-classification-system