IP Report 2019

Trade marks: is Australia's register cluttered?

Trade mark cluttering occurs when a trade mark register contains many unused or overly broad marks that block the same or similar marks from being registered. Cluttering of a register increases the costs to other applicants of creating and registering new trade marks.1

The 2016 Productivity Commission inquiry into Australia’s Intellectual Property Arrangements highlighted trade mark cluttering as a potential problem in Australia. In responding to the Commission’s report, the Australian Government asked IP Australia ‘to undertake further research and analysis to determine the sources and extent of any clutter on the trade marks register’ before developing potential reforms.2

In investigating this issue, IP Australia’s Office of the Chief Economist (OCE) examined the available evidence, consulting externally through academic seminars and internally, to determine the extent of trade mark cluttering in Australia.

Trade mark cluttering in Australia is not an immediate concern

Measuring trade mark cluttering is challenging as it is difficult to know if a trade mark is still in commercial use or if it is unjustifiably blocking use by others. The OCE’s research creates and develops a new and comprehensive set of measures to probe the trade mark cluttering situation in Australia. Overall, our results suggest that there is a low incidence of cluttering in IP Australia’s trade mark register, although there is evidence of applicant behaviour that can lead to additional trade mark cluttering in the future.

A very small number of marks are removed due to non-use

The Australian Trade Mark Act (1995) allows a third party to challenge and remove another person’s trade mark under certain conditions where a mark is not being used.3 When someone is petitioning to remove a trade mark from the register because it is not being used, we considered this an indicator of possible clutter.

Figure 17 illustrates that although the number of marks removed due to non-use doubled, from 219 in 2006 to 437 in 2016, these represent less than 0.1 per cent of the total marks in force each year. This proportion has not increased over the last 10 years, indicating that the increased number of non-use marks which are actively removed for being unused does not represent a deterioration of the Australian trade mark register.

Figure 17 is a graph of the number of trade marks removed due to non-use and the total number of marks in force from 2006 to 2016

A potential source of trade mark cluttering has been growing

Once trade marks are registered they are automatically valid for the first ten years and can be renewed every ten years. The trade marks not renewed after ten years could arguably be identified as non-use trade marks. While we know that the mark was not commercially useful in the eleventh year, it might also have been unused in the tenth or ninth year. It is almost impossible, though, to know exactly when a mark stopped being used.

The renewal rate after the first ten-year registration term has dropped from around 70 per cent in the 1980s to 50 per cent in the 2000s (Figure 18). This has occurred during a period of strong growth in trade mark registrations within Australia, the total marks in force rising by more than 60 per cent, from 391 450 in 2005 to 635 355 in 2017.4 The number of potential non-use trade marks which remain on the register during their first ten years has been increasing, and this has likely contributed to trade mark cluttering in Australia.

Figure 18 is a graph showing first renewal rate by trade mark filing year from 1980 to 2006

Other indicators also support our key finding

A number of indirect indicators of trade mark cluttering are also included in our research. The first, inspired by work in the US, is a study of single-word marks for popular words in the registers of the United States Patents and Trademarks Office (USPTO) and IP Australia. In both countries there are lists of the most popularly used words, and we compare these to the trade mark registers in each country to see how ‘full’ the register is.

These popular words indicate demand for trade marks as applicants seek concise, easily understood marks to support their brand. Our analysis finds that there are fewer single-word trade marks registered in Australia than in the US. For example, of the most popular 1 000 words, 81 per cent are registered in the US but only 56 per cent are registered in Australia (see Table 2). This may indicate that the Australian trade mark register is less crowded than its US equivalent, or that the relative value of a market niche is greater in the US than in Australia.

 Table shows proportion of most frequently used words matching active single-word marks.

Despite the data indicating a low level of clutter, the average number of words registered in an Australian trade mark has been increasing, from below 1.5 before 1980 to above two after 1990. It appears that new applicants are filing marks that contain more words and are more complex than single-word marks. A possible reason for this change in behaviour is that popular single-word marks are already in the register, precluding their registration More complex marks may be less effective to applicants, as research indicates that concise trade marks are more valuable to their owners.5

Another aspect of clutter is the scope of protection, where a trade mark is registered in NICE classes which are not being utilised in the marketplace. Our analysis did not find strong evidence of trade mark cluttering in any particular class of trade marks in Australia, so this did not seem to be an issue.

In summary, we find that at present the overall state of trade mark cluttering in Australia is not unduly hindering the system. The number of ‘cluttering marks’ has been increasing, but their ratio with the total stock of active trade marks is still small. It appears the current mechanisms for removing non-use marks that block other traders are working effectively. The research paper is available as an economics working paper on the IP Australia website at www.ipaustralia.gov.au/economics


TM-Link is a single, internationally linked trade mark database, allowing researchers, businesses and IP offices to track the use of similar trade marks across different countries. IP Australia, in collaboration with The University of Melbourne and Swinburne University of Technology, built the database and released its beta version in 2017.

Despite TM-Link still being in its infancy, it is already amassing a substantial amount of data: the database currently includes data on 15.3 million trade mark applications filed in Australia, Canada, the European Union, New Zealand, the UK and the US. These applications were filed by over five million applicants from 238 unique geographic regions. Over the coming years, the data is expected to grow, as more IP offices add their trade marks to the database.

IP Data Platform

The IP Data Platform is an exciting new initiative that has just been released by IP Australia.

The IP Data Platform will provide registered users with access to a cloud-based analytics lab. The lab contains data science tools that support users to interrogate the TM-Link database, upload their own data and collaborate with other users. For governments, universities and businesses worldwide, the benefits from using TM-Link via the IP Data Platform will be far-reaching.

Analytical insights from TM-Link

TM-Link data can be used to understand patterns of international trade mark filing activity. As a preliminary example, the Office of the Chief Economist (OCE) looked at the foreign-origin trade marks of four countries—the US, Canada, the UK and Australia—where full data was available, and compared applications in each class across these countries.

Of the four countries, Australians file relatively more trade marks overseas in the Wines and spirits class than do applicants in other countries. Toys and sporting goods, as well as Hotels and restaurants, also feature prominently in Australian trade mark applications filed overseas.

Results were starkly different for the other three countries. Textiles, Leather goods and Telecommunications top the list in UK-origin applications filed overseas; Building materials, Construction and Treatment of materials were the top classes in applications by Canadians; and Firearms, Lubricants and Fuels and paints were the top classes in applications by US applicants.

As more countries’ data are included in TM-Link, it will be possible to profile their international trade mark filing activity for the purpose of generating policy insights.

Supporting the development of TM-Link and the IP Data Platform

IP Australia is pleased to offer free trial access to the IP Data Platform to researchers, data scientists, developers and other interested parties in the broader IP community. Participants will be invited to share their work in order to encourage greater collaboration. As TM-Link and the IP Data Platform continue to be developed and enhanced, IP Australia is inviting interested parties to participate in a free trial and contribute to its design.


View the IP Report 2019 in PDF Format

End notes

  1. von Graevenitz, G., Greenhalgh, C., Helmers, C., & Schautschick, P. 2012. Trade Mark Cluttering: An Exploratory Report. London: UK Intellectual Property Office.
  2. Australian Government. 2017. Australian Government Response to the Productivity Commission’s Inquiry into Intellectual Property (IP) Arrangements. Canberra: Commonwealth of Australia, viewed January 2019.
  3. These conditions included the situation where five years had lapsed since the mark’s filing date and where the mark had been registered for a continuous period of three years without being used.
  4. WIPO IP Statistics Data Center 2018, retrieved January 2019.
  5. Beebe, B. & Fromer, J.C. 2018. Are we running out of trademarks? An empirical study of trademark depletion and congestion. Harvard Law Review, 131(4); 966–67.