Quantum technologies: patent applications vs. scientific publications across the world

By Sébastien Ragot and Michel Kurek

A second quantum revolution is underway, which promises to revolutionize several fields of technology. The outlook is exciting, but some indicators are disturbing. In particular, there are glaring differences in the ratios of patent applications to scientific publications between countries. The situation is particularly worrying for European countries, which may pose questions in terms of economical sovereignty in the long run. However, there are also reasons to be optimistic. Plus, it is not too late to adequately harness patents, given that such technologies will not deliver their full potential before years.

A new quantum revolution is in motion, driven by recent advances in quantum information science. Various fields are being conquered, which notably concern quantum computing, quantum communications, and quantum sensing. Quantum computing targets applications in information processing; the hope is to largely outperform classical computers, at least for certain tasks. Quantum communication aims at securing communications and quantum sensing seeks to beat limits in classical sensor technology. These new fields have in common the exploitation of quantum phenomena such as quantum teleportation, quantum entanglement, quantum interference, and quantum state squeezing. Such phenomena have no classical equivalents and are thus difficult to comprehend—even for physicists.

The quantum race began quietly in the 2000s and has accelerated markedly since the 2010s. Such technologies are expected to have a major impact on society and the economy. This is why governments have rolled out strategic plans with substantial funding commitments, in the millions and billions, depending on country size and ambition. 2020 was a record year for private investment in the space and 2021 looks set to deliver as much capital as the previous three years combined. Major ICT players (such as IBM and Google) complete the investment landscape with development strategies that heavily rely on R&D and communication.

Such influxes logically translate into start-up creation, scientific papers, and patents. More than 34,000 scientific publications related to quantum technologies can be identified in the period 2010 – 2020 (against 43,000 in 2001 – 2021), with an average annual growth of approximately 11 % (up to 20 % in recent years). 81 % of these arise from just 10 countries. Interestingly, Europe accounts for the highest share of publications (32%, down to 23% looking solely at EU countries, mainly due to the UK), ahead of the US (21 %) and China (19 %).

Figure 1: Breakdown of scientific publications by country or region of origin over the period 2010 – 2020

What about patents? One would logically expect to see a similar pattern in the published patent applications as in scientific publications. However, this is far from the case.

More than 18,000 patent applications have been published in the field of quantum technologies since 2000, but the vast majority (more than 15,000) were published after 2010. Figure 2 depicts the distribution of patent applications published since 2010 in all jurisdictions, by country or region of origin. Clearly, China and the US hold the upper hand. The two countries are responsible for nearly three-fourths of published patent applications and all of the top 10 applicants (6 Chinese and 4 US applicants). Broadly speaking, the US still dominates the patent game in the quantum computing area, while China has taken a clear advantage in quantum communications.

Figure 2: Breakdown of patent applications (quantum technologies) published since 2010 by country?or region of origin (all jurisdictions)

Asia also accounts for the majority of patent applications: China, Japan, and South Korea together account for 60% of the patent applications published worldwide. However, this last figure is at odd with the trends observed for scientific publications, where the combined contributions of China, Japan, and Korea fall to just 25% of published papers, whereas the US and Europe contribute 53%. In detail, the patent application share (23%) of the US is consistent with its scientific publications (21%), while the European contribution to published patent applications is quite disappointing (8% only, down to 5% for the EU) compared to its share of scientific publications (23–32%). On the contrary, the Chinese patent share (50%) markedly exceeds its contribution to the scientific literature (19 %).

There are several phenomena at work behind such numbers, notably related to culture, financial investment capacity, entrepreneurship, and local incentives. For instance, many European companies favor trade secrets over patents and most European scientific publications arise from the academic sector, which is traditionally less concerned with patents than the private sector. Moreover, European businesses have been in decline (relative to other parts of the world) since 2008, which has impacted their financial investment capacity.

The discrepancies noted above in respect of ratios of patents to scientific publications may be interpreted in various manners. One may think that some applicants underexploit the patent system, while others overexploit it, or at least utilize it to an extent exceeding expectations based on scientific publications. Such discrepancies may also mean that some applicants regard quantum technologies as being less strategic than others, right or wrong. Also, these discrepancies may, at times, call into question the true technical contributions of some patent applications and therefore the patent quality.

Still, it should be remembered that Figure 2 concerns published patent applications (worldwide), and not granted patent applications. That is, not all of these patent applications have been or will eventually be granted by the patent offices. In that regard, one may consider using grant rates as a measure of patent quality. However, grant rates are misleading because patent investments by companies differ from one jurisdiction to the other and patent offices have fairly different grant rates, let alone differences in terms of search and examination practices.

Several proxies and metrics can be used to assess patent quality, such as forward citations, originality, radicalness, technical impact, and other pointers as routinely provided by professional patent databases. Such metrics tend to show that patent applications from US and UK applicants are the most robust and have the most technological impact. US-originated patent applications would also be more original, compared to European and Chinese applications. However, such metrics must be interpreted with caution. For example, those based on forward citation counts can be misleading because some examiners tend to more systematically cite documents that they already know, it being noted that the relevant technical literature was initially dominated by the US. Plus, such metrics do not take into account the various “patent cultures” in the world.

Figure 3 shows the breakdown of patent applications published since 2010 by country or region of origin, in three major patent jurisdictions: the European Patent Office (EPO), the US patent office (USPTO), and the Chinese patent office (CNIPA). The resulting diagrams make it clear that Chinese applicants focus on China, while only marginally filing in Europe and the US. They further confirm that European countries are outclassed, even in their own jurisdiction (EPO). This, in the long run, may pose questions in terms of economic sovereignty.

Figure 3: Breakdown of patent applications (quantum technologies) published since 2010 by country or region of origin in the three main patent jurisdictions (CNIPA, USPTO, and EPO)

Besides the quality metrics evoked earlier, other proxies that can be used to assess patent quality are counterpart ratios, here calculated as the number of patent applications published in the two foreign jurisdictions (e.g., CNIPA and EPO) compared to the numbers of patent applications published in the jurisdiction of origin (e.g., USPTO). Such ratios are now much more favorable to European-originated patent applications (25-110%), than US-originated patent applications (15-24%) and Chinese applications (1-2%). However, such ratios do not take into account national filings in European countries, which are about as numerous as EPO filings. In addition, they are impacted by several factors such as local incentives to file, language barriers, and culture. All the more, these ratios actually reflect the capacities, strategies, and ambitions of the applicants, who decide to extend patents (or not) based on their appreciation of both the patentability and the business incentive to file. ??

To summarize, the discrepancies observed in respect of ratios of patent applications to scientific publications reveal differences of strategy and/or capacity. They may further put into question the technical contributions of the patent literature, something that is difficult to assess with usual quality metrics. Chinese applicants massively file patent applications, to an extent that substantially surpasses their share of scientific publications, but essentially target China. US applicants dominate in their jurisdiction and invest in both European and Chinese patents. The relative share of patent applications published by US applicants is on par with their scientific papers. Finally, Europeans holds the highest share of scientific publications but are overwhelmed in terms of numbers of patent applications published. Still, they seem quite confident about their inventions, which they readily extend to the other two jurisdictions.

That said, it is not too late for the Europeans (and others) to straighten things out, given that quantum technologies will not deliver their full potential for years to come. There is ample room for further technical progress and, therefore, for many more patent applications. Being fundamentally rooted in hardware, quantum innovations are mostly patent-eligible, which makes them a priori easier to patent than other trendy technologies, such as blockchain and machine learning. Yet, there are a few pitfalls that patent applicants need to be aware of, as discussed in detail here.

Thus, many more patent applications will likely be filed in the quantum area in the coming years. And many more scientific papers will be published too. Excesses growth in patents compared to scientific publications may be a sign of disquieting differences emerging in players’ strategic positioning. Or may simply call for caution with respect to the true technical contribution and/or quality of the patents concerned.???