China's Spies Elude U.S. Vacuum Cleaner
Industrial espionage is too complex to be captured by a pat metaphor.
President Donald Trump could never be accused of underestimating the impact of Chinese economic espionage and technology transfer on the United States. “We’re talking about big damages,” he said when discussing retaliation for intellectual property theft in a January interview. “We’re talking about numbers that you haven’t even thought about.”
His tariffs and a recently floated proposal to restrict certain Chinese researchers in the U.S. are calibrated to be equally tough. But will they work?
To answer this question, I spoke to Mara Hvistendahl, a national fellow at the New America Foundation, a contributing correspondent at Science, and an expert on Chinese industrial espionage. During the eight years she covered science and politics from China, she wrote a series of groundbreaking stories that demystified and personalized China’s hackers and scientists. She’s currently at work on her next book, “The Scientist and the Spy,” which examines industrial espionage, China, and the FBI. It’ll be published in 2019. Our conversation has been lightly edited:
Adam Minter: In February FBI Director Christopher Wray told a Senate Intelligence Committee hearing that Chinese students, academics and researchers are exploiting the U.S.’s “very open research and development environment” and constitute a “whole-of-society threat.” Is that a new outlook at the bureau?
Mara Hvistendahl: To be clear, industrial espionage — which is sometimes perpetrated by Chinese professors and researchers — is a significant threat. But Wray’s comments touched a nerve because they follow on a long line of intelligence officials making blanket statements about the role played by Chinese-Americans in Beijing’s intelligence efforts. For years, the prevailing theory about Chinese espionage was that Beijing used a “vacuum cleaner” approach, relying on a large number of amateur collectors of Chinese descent to vacuum up small amounts of information. You can see how that metaphor lends itself to broad generalizations about an entire group. But it turns out that the metaphor is wrong.
The vacuum-cleaner theory — which dates back to the late 1990s — has lately come under attack by a new cohort of experts on Chinese intelligence, most notably Peter Mattis, a research fellow at the Victims of Communism Memorial Foundation. If you look at the actual espionage cases that have emerged from China, there is a broad diversity of approaches taken. Instead of one central actor controlling an army of informal collectors, a number of different actors are often separately pursuing a variety of goals. So spying might originate with the Ministry of State Security or the People’s Liberation Army, or it might originate with private-sector businesses that have other motives for wanting information. And these entities often rely on collectors who are not of Chinese descent. But this more complex landscape is much harder to sum up with a pat metaphor. So the vacuum-cleaner comparison won’t die.
AM: To what extent are China’s industrial policies dependent on information gleaned from foreign research efforts, including universities, whether legally or illicitly?
MH: If you look at Chinese development plans like Made in China 2025, which aims to make China a leader in strategic areas like robotics and advanced medical devices, there’s a clear emphasis on assimilating foreign technologies. What makes this issue so complicated is that this emphasis exists alongside legitimate and robust scientific development. There is now extensive scientific cooperation between the United States and China, with many researchers traveling back and forth and often even holding posts in both countries. The FBI now has to make sense of that. And it’s not easy.
Scientists, meanwhile, are caught in the middle. On the one hand, they are often encouraged — by their universities, by U.S. government research organizations — to collaborate with colleagues in China. On the other hand, they worry that they could be prosecuted for a wrong move. So they’re getting mixed messages.
AM: One of the themes in your work on Chinese hacking is that it’s wrong to assume every incident or plan is cooked up at the highest levels of China’s intelligence bureaucracy. Does it go for economic espionage?
MH: Yes. There have certainly been examples of Chinese military or state ministry personnel engaging in industrial espionage, but these typically involve technologies with military applications. When a Chinese business sets out to steal a U.S. competitor’s technology or trade secret, state approval is often implicit rather than explicit — meaning that the Chinese government prioritizes certain areas for development, and then looks the other way once an abuse is committed.
Policymakers often get hung up on whether the Chinese government is directly involved in a specific case and end up overlooking the role played by the private sector. Some of the experts I talked with point out that to really combat IP theft, you have to create penalties that affect the companies responsible for it, rather than just going after the researchers who are caught in the act of stealing a trade secret. To compare this to the fight against drug trafficking, you can’t go after drug runners without looking at the cartels.
AM: In recent years, the FBI has successfully prosecuted several cases of Chinese economic espionage, including some connected to research institutions. But there have also been some very high-profile failures that raise questions of outright prejudice, as well as agent and prosecutorial incompetence. Is there a risk that — as the FBI raises the profile and fears around such cases — we might see more of the latter than the former?
MH: Any time something becomes a top priority, you see questionable cases being brought. This happened with terrorism after 9/11 and it’s happening now with industrial espionage.
But groups like the Committee of 100 point out that after 9/11, the FBI reached out to Muslim American groups to ensure that the bureau’s response was appropriate. They’re calling for the same thing to happen now. Scientists need to be turned into allies rather than potential threats in the effort to combat industrial espionage.
It’s also important to note that the long-term fallout from a single bungled case can be significant. If you look at the Wen Ho Lee case, for example: In 1999, Lee was accused of stealing secrets connected to the U.S. nuclear arsenal. He committed clear security breaches, but the investigation into his actions spun out of control, leading activists to allege that Lee was targeted because of his race. Not surprisingly, this perception damaged recruitment efforts at America’s leading weapons labs. In the years following the debacle, Los Alamos had difficulty attracting new scientists of Asian descent. Several Chinese-born scientists who worked at the lab, meanwhile, accepted positions back in China. So many left, according to a South China Morning Post report, that a group of researchers in China became known as the “Los Alamos club.”
AM: Even before the Trump administration, there seemed to be fairly widespread, bipartisan, even international, support for the idea that China’s technology transfer demands had gone too far. What was the preferred route to dealing with them before the Trump administration?
MH: There have been a number of approaches put on the table over the past five years, most notably by the IP Commission, an independent body then led by Jon Huntsman, the former U.S. ambassador to China (and now U.S. ambassador to Russia). These include blocking offending companies from using the American banking system and improving the International Trade Commission’s process for sequestering goods containing stolen IP.
The Obama administration made a lot of noise about IP theft from China, and had some limited success in addressing the issue. So President Trump is certainly not the first president to take on this challenge.
AM: The Trump administration raised the possibility of excluding certain Chinese researchers from the United States, or at least restricting their access to various research areas. What impact would that have on U.S. science?
MH: Additional security reviews in some fields might make sense. But reducing the total number of visas granted — if indeed that is what’s planned — would be a mistake. There’s no question that Chinese scientists and engineers are critical to innovation. According to the National Science Foundation, about half of the Ph.D.s in science and engineering granted in the United States each year go to foreigners: of these, almost a third are Chinese citizens, more than to any other nationality.
The IP Commission actually recommended increasing increase the number of green cards given to Chinese students in the United States, to induce them to stay rather than return to China. And that makes sense to me. We need to proceed cautiously. But if the end goal in cracking down on industrial espionage is to safeguard American innovation, we’re not going to accomplish that by closing off our borders to the people who power our scientific community.
This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.
To contact the editor responsible for this story:
Matthew Brooker at firstname.lastname@example.org