Home / HBD / China Overtakes US in Scientific Articles, Robots, Supercomputers

China Overtakes US in Scientific Articles, Robots, Supercomputers

Nothing illustrates China’s meteoric rise as some well chosen numbers.

By the end of the 1990s, China had come to dominate the mainstays of geopolitical power in the 20th century – coal and steel production. As a consequence, it leapt to the top of the Compositive Index of National Capability, which uses military expenditure, military personnel, energy consumption, iron and steel production, urban population, and total population as a proxy of national power. Still, one could legitimately argue that all of these factors are hardly relevant today. While Germany’s fourfold preponderance in steel production over Russia may have been a critical number in 1914, China’s eightfold advantage in steel production over the US by 2014 is all but meaningless in any relevant comparison of national power. The world has moved on.

By the end of the 2000s, like Victorian Britain in the mid-19th century, China became the workshop of the world, overtaking the US in both manufacturing and coming very close to it in terms of PPP-adjusted GDP. As a consequence, this was when China also overtook the US on a wide range of consumer welfare and ecological impact indicators, such as exports, CO2 emissions, Internet users, energy consumption, car sales, car production, and number of patents issued. Still, its presence in the hi-tech sector was still pretty modest, and innovation was low. This was not yet an economy that could furnish first-class armaments, or inspire far off peoples to carry out color revolutions in its name.

But as of this year, China is hurtling past yet another set of inflection points – the hi-tech component of its economy, roughly comparable to any of the major European Powers a mere decade ago, is now about to converge and then hurtle past that of the US by the end of the 2010s (even if in per capita terms it remains considerably behind, like South Korea 20 years ago).

This process can be proxied by three indicators: Number of scientific articles published, operational stock of industrial robots, and number of supercomputers.

Science Articles

The SJR maintains a database of scientific publications by country and subject for the past 20 years.

In 1996, China published a mere 29,000 papers, well behind Japan, the UK, Germany, and France (50,000-90,000) not to mention the US with 333,000. As of 2015, however, China had surged to 416,000 published papers, still modestly behind the US with its 567,000 papers but far ahead of everyone else.

Now to be sure, Chinese papers are still considerably less cited than those of the developed world. And yes, this reflects the science-plagiarism-mapfact that, on average, the quality of Chinese scientific output remains inferior – less innovative, more derivative – than that of the US. This extends to outright plagiarism; the negative stereotypes about Chinese academia are somewhat borne out by a study that showed that 7-8% of Chinese articles on arXiv.org were flagged for text overlaps, compared to less than 4% for the US and the UK.

Nonetheless, in the “hard”/STEM spheres that arguably matter more for technological progress – and which have much less in the way of a replicability crisis – China is already ahead of the US in terms of total publications: 34,000 to 28,000 in mathematics; 67,000 to 52,000 in physics and astronomy; 63,000 to 36,000 in chemistry; 120,000 to 67,000 in engineering; 49,000 to 41,000 in computer science. The only major spheres here in which the US remains considerably ahead are the more biologically orientated sciences, such as: 196,000 to 69,000 in medicine, 83,000 to 59,000 in biochemistry/genetics, 23,000 to 7,000 in neuroscience, and 18,000 to 14,000 in pharmacology. Otherwise, the US retains clear dominance only in the the softer spheres of social science and the arts: 54,000 to 7,000 in the social sciences, 10,000 to 2,000 in economics, 23,000 to 2,000 in psychology, and 27,000 to 2,000 in the arts and humanities. In one subcomponent that is arguably outright negative value added, that of Gender Studies, the US published 1,456 documents to China’s 23.

The overall trends cannot be denied – Chinese scientific output is rapidly approaching American levels and will probably outright overtake, at least in absolute numbers, by around 2020.


Until recently, the general consensus was that automation would be an issue mainly for developed countries with high labor costs. China, then still seen as a country of boundless, cheap, and disciplined if unskilled labor, was not expected to be deeply affected by those developments (except perhaps to the extent that it would be challenged by renewed competition with First World manufacturing “reshoring” back to the American rustbelts).

This was, until recently, a logical enough viewpoint. Traditionally, the world’s operational stock of industrial robots was concentrated in the most advanced manufacturing economies, with the highest per capita rates seen in Japan (which accounted for a third to half of all industrial robots during the 1980s and 1990s), Germany and the Germanic lands, Northern Italy, and more recently, South Korea. In contrast, until the early 2000s, the publicly available databases generally didn’t even bother to estimate the numbers of industrial robots in Chinese factories so small and insignificant were their numbers.

But from the late 2000s, the robotization of Chinese industry began to explode.


China went from having 32,000 industrial robots in 2008 (~Spain), to 189,000 by 2014 (~Germany) and approximately 263,000 robots by 2015, which puts it ahead of the 259,000 robots in all of North America and just behind Japan’s 297,000. It is therefore safe to assume that China took first place this year. By 2018, China is projected to have 614,000 industrial robots, equal to that of Japan and North America combined.

It is also worth noting that China dominates the global machine tool production industry, having overtaken the two leading countries in that sphere – Germany and Japan – around 2010. As of 2014, China accounted for 30% of the world’s yearly production of machine tools. This is of special interest not only because of this industry’s inherent technological sophistication, but also because of its strategic importance as the only part of the industrial economy that actually reproduces itself and makes everything else possible.


A third excellent proxy for a country’s technological sophistication is its stock of supercomputers, which enable detailed simulations of phenomena as disparate as global climate, protein folding, and nuclear weapons reliability.

China emerged on the supercomputing scene in force during the early 2010s, when it became the world’s (distant) second to the US. However, within the space of the past year, it has surged ahead. According to the June 2016 list of the world’s top 500 supercomputers, China is now marginally ahead of the US in terms of total number of systems, with 168 top systems relative to America’s 165, and well ahead in terms of performance share, with 211 petaflops total to America’s 173 petaflops.


China also hosts the world’s most powerful single supercomputer, the Sunway TaihuLight, which is nearly three times as powerful as the world’s second best (also Chinese) and five times as powerful as the top US supercomputer. Remarkably, it is based entirely on Chinese processors, the US having banned the export of Intel chips used in previous Chinese supercomputers for national security reasons in 2015. Evidently, this has had negligible effects on Chinese technological progress, because China has no dearth of native human capital and a state-backed program to reduce reliance on foreign technologies.


Forget the war against terror, forget the Syrian conflict, forget Ukraine – when historians look back on this period, they will identify China’s emergence as a technologically capable continental economy (soon to far overtake the US in absolute size) that is less and less reliant on the West for its technological convergence is by far the most important geopolitical trend of the century.

As this process unfolds, China is likely to start being more assertive on the international stage. We are already seeing this in the South China Sea, and its recent aquisition of its first foreign military base in Djibouti and plans to multiply its (as yet meager) power projection capabilities by building over 1,000 heavy strategic aircraft – that’s far more than what the US and Russia have combined. (Note that my standing projection is for China to overtake the US in total military power by 2030 and in naval power by around 2040).

It will also come to assume a much bigger presence in science, culture, and soft power generally, though this will take some time to recognize given the long lag times between invention and recognition.

Its also worth emphasizing that this technological emergence is quite specific to China, not to the BRICS in general. South Africa is basically an affirmative action BRIC and not worth mentioning further, while Brazil is the country of the future – and always will be, as per De Gaulle’s witticism. Despite strong recent economic growth, India’s presence in all the aforementioned spheres – published papers, supercomputers, industrial robot stock – is comparable to that of a typical middle-sized European country, its huge population being nullified by underdevelopment and an average national IQ in the low 80s.

As for Russia, while general economic output has recovered and exceeded Soviet era levels, its scientific and technological superstructure remains depressed: Russia’s share of global science papers as of 2015 is now 1.9% of the world’s total relative to 7.6% in 1986 (a drop made all the more remarkable by the USSR’s absence of a “publish or perish” scientific culture); its respectable Soviet-era stock of ~60,000 industrial robots has now almost entirely depreciated without getting replaced; and the quantity of Russian supercomputers in the top 500 in any given year has stabilized at around 5-10 since the late 2000s (i.e., comparable to Sweden). This is a consequence of the post-Soviet degradation of Russia’s human capital, especially its more elite elements, due to the 1990s brain drain; the ultimately lackadaisical approach to industrial and technological policy under Putin; and the intrinsic limitations of a ~97 average national IQ (in comparison, China, Germany, Japan, and the advanced parts of the US and Italy are in the low 100s).