Getting Chippy: Export Control Advocates and the Endless Sanctions Logic Trap

A recent article from Noah Smith illustrates a disconcerting parallel between supporters of high-tech export controls and advocates of endless, broad-based economic sanctions policies

If you believe a policy is good and necessary, you should be able to offer a better defense than saying “temper your expectations” for the policy and calling its opponents unrealistic.

I have written previously about the Biden Administration’s October 2022 export controls which restrict export of certain advanced node semiconductors and related equipment for manufacturing these advanced devices to China. My feeling is that they may end up doing more harm than good. Not that proponents of the controls don’t have valid reasons they can point to.

Civil-military fusion in China does make it difficult to discriminate end uses of advanced semiconductor technologies, and some degree of anxiety about military applications of artificial intelligence which rely on advanced semiconductors is warranted. In particular, autonomous “drone swarms,” a theorized military technology where groups of drones act in concert with one another like a flock of birds, has the potential to be an effective future weapons system.

A proposed future military application of advanced logic chips is autonomous drone “swarms” powered by AI, acting in concert like this beautiful flock of geese. Photo by Gary Bendig on Unsplash.

In fact, the Pentagon recently announced the “Replicator Initiative,” which has the ambitious goal of fielding “attritable autonomous systems at a scale of multiple thousands, in multiple domains, within the next 18-to-24 months.” Clearly the United States sees artificial intelligence applications as critical and potentially decisive in future conflicts.

However, the cost to enacting these controls is potentially high and the benefits far from certain. Some experts have warned the controls are doomed to failure, and even defenders of the controls acknowledge they are tantamount to a declaration of economic war on China.

I worry that the logic at play in arguing for the controls parallels a similar, dangerous logic: that of hawks arguing for broad-based economic sanctions like those imposed by the United States on Iran and Venezuela. These sanctions seldom produce the desired outcome, often remain in place indefinitely, and their advocates say tightening the screws just a bit further, pushing a bit harder, will lead to success.

I see a very similar line of reasoning in a recent article by economics writer Noah Smith. I will explore this further below, but first we must discuss the relevant news.

Last week, much was made about Huawei’s release of the Mate 60 Pro, a smartphone which uses a 7-nanometer processor, the Kirin 9000S. For reference, the Biden Administration’s advanced semiconductor export controls apply to all technologies below the 14- or 16-nanometer processing nodes. The 7-nanometer processor produced by the Chinese semiconductor manufacturing company SMIC puts the company nearly on par with Intel and Samsung, according to this piece from Semianalysis. The release of this phone represents a breakthrough in native Chinese semiconductor production capabilities, and according to some, a failure of the United States semiconductor export controls.

The economics writer Noah Smith says not so fast. He released a piece, “Who's afraid of the Huawei Mate 60 Pro?” which argues that those lamenting the failure of the controls may be speaking too soon. Noah is a strong supporter of the export control regime, and while he often provides solid economic analysis, I think he misses the mark here. He says the argument that export controls will only push China to innovate faster “makes little sense,” since they were already highly motivated to build a domestic semiconductor industry.

However, this entire piece is downplaying just such an accelerated innovation Huawei accomplished while under U.S. sanction. A knowledgeable commenter on this article, Jan, says:

When Huawei was first added to the Entity List in 2019 and then restrictions were tightened further in 2020 to preclude its use of TSMC to fab its in-house HiSilicon chips, the state of the art was 7nm (with 5nm in sight). Lots of observation at the time pegged China as being "roughly a decade" away from being able to reach best in class. And yet, 3-4 years later, they've now reached what was (at that time) the best in class.

This has taken them less than half the time many predicted and demonstrates not so much the power of Beijing's industrial policy, but the power of what can be achieved when the incentives of China's leading private sector players (e.g., Huawei) are forcibly aligned with the Chinese state's due to US export controls. Huawei was happy to continue procuring from US suppliers (it was buying ~$20B a year!). Yes, it was developing in-house alternatives for many components as part of its longstanding strategy of vertical integration, but it saw its central role in global tech supply chains as something that was desirable, not regrettable. Until the US government intervened.

Jan adeptly illustrates precisely how U.S. sanctions can inadvertently accelerate Chinese innovation by aligning state and private incentives, and isolating companies from the global supply chain. This represents a major risk in the Biden Administration’s policy targeting the entire advanced semiconductor industry in China.

Jan takes particular issue with one line in Noah’s piece, which also stood out to me. Noah says:

It’s hilariously unrealistic to think that if the Biden administration dropped all export controls tomorrow, Chinese companies would just go right back to buying their chips from Qualcomm, cheerfully relying on U.S. technology and scrapping their plans for innovation.

Jan responds:

The above quote belies a complete misunderstanding of the current situation in China. All of Huawei's main competitors (Xiaomi, Oppo/Vivo/OnePlus, Lenovo, ZTE, etc.) procure most of their chips from US suppliers, particularly Qualcomm. That's because they've wanted to and they have to: not only is Qualcomm best in class, its main alternative (MediaTek) isn't a strong competitor at the high end. Those firms will all continue on procuring from Qualcomm (and Intel and Broadcom and Qorvo, and Skyworks, etc., etc.) if they are allowed. But unlike a year ago, if the US government decides to export control those Chinese firms too, they have a credible domestically-sourced alternative: Huawei's chips fabbed at SMIC. Based on rumors trickling out of the China tech scene, the same may soon be true for HPC [high-performance computing] needs and Nvidia. The closing of these gaps doesn't get China all the way to the cutting edge, but it buys it time. We shouldn't act like they aren't in parallel frantically pursuing a domestic EUV capability - various research and patent filings are giving glimpses of the movement on that front.

Yes, Beijing's march for self-sufficiency would have continued in either case. But Washington's moves against Huawei inadvertently (though predictably) pumped rocket fuel into that effort in a way that is already having significant unintended effects. Hand-waving away any potential for a modification of US policy to re-calibrate private sector incentives in China is a mistake. Washington still has choices.

I could not do a better job addressing the substantive issues with Smith’s rather glib dismissal of opponents of the Administration’s sanctions policy, so I will let the above speak for itself. I will add, however, a critique of the form of Noah’s argument, and question some of its underlying assumptions.

I would like to note how eerily his recommendation parallels the arguments of some of the most hawkish advocates for endless, broad-based economic sanctions, such as those currently levied against Venezuela and Iran.

Underlying Smith’s support for the advanced semiconductor controls is the assumption that the United States and China are on a collision course, so the United States should act to attempt to degrade China’s military preparedness. The possibility of conflict, by this logic, practically demands action.

Putting aside the countless non-military uses of advanced semiconductors, and the fact that many of the most fearsome military technologies advanced chips may enable do not even exist yet, this attitude of inevitability is dangerous. It insists on continuing escalation by the United States, making alternative courses out to be “hilariously unrealistic.”

Daniel Larison, author of the excellent newsletter Eunomia and perhaps the sharpest critic of broad-based economic sanctions, describes this phenomenon here:

Michael Beckley has written a terrible article about the supposed perils of détente with China. It’s a familiar story that hawks love to tell every time that the U.S. finds itself on a collision course with another country. As they see it, rivalry is baked in and inevitable because of irreconcilable differences, and attempts to lessen tensions are seen as either useless or harmful because they play into the adversary’s hands.

This frees hawks of any responsibility for having put the U.S. on the collision course through the policy choices that they cheered on, and it allows them to dismiss the possibility that relations might be improved through negotiations and compromise. When you make it seem as if there is no avoiding rivalry and no way out of it, it makes it much easier to sell hardline policies. According to this extremely convenient worldview, confrontational U.S. policies and actions have nothing to do with creating and stoking the rivalry, but when the U.S. offers olive branches and tries to lower the temperature in the relationship it just encourages the other side to press its advantage. It should be obvious that this way of looking at the world is total nonsense.

While it is important to prepare for the worst in international relations, it is also important to recognize that what the United States does and says matters, and escalatory actions by the U.S. are a choice, not inevitable. In the chips case, the United States actively chose to pursue a provocative policy deliberately designed to weaken an entire high-technology industry in China. They did this, as Smith acknowledges in the quote below, without knowing whether or not it will actually impact military modernization, the stated goal of the policy:

What effect all this will have on China’s military capabilities isn’t clear, because that depends a lot on the pace of military innovation. It might be that old chips are fine, and China can just overwhelm the forces of developed democracies with sheer mass of missiles and drones. Or it might be that new AI algorithms that allow the creation of highly effective drone swarms might need to run on bleeding-edge chips, in which case China could be at a qualitative disadvantage in a fight. We just don’t know for sure; the argument that China’s military will be hampered by a weaker chipmaking industry is probabilistic, not a law of the Universe. But it’s still one thing that might help put a check on Chinese expansionism in the 2020s and even the 2030s, and every little thing counts.

Note the framing: the U.S. is simply acting to “check” “Chinese expansionism”. A future conflict is all but a sure thing. The United States’ motivation is all very selfless and logical, not an aggressive push to maintain U.S. and allied dominance over a key high-technology area with critical economic and technological importance. Something tells me this would not be the framing if China acted to strangle some American industry which China currently dominates, like electric batteries or solar cells. In that case China would rightly be seen as aggressively wielding its market dominance in a manipulative way in order to gain an unfair competitive advantage.

Finally, we get to the quote from Smith’s piece which inspired this essay. The following recommendation has been repeated over and over, nearly word for word, by advocates of broad-based economic sanctions which have had devastating effects on millions of innocent people around the world while producing few tangible benefits. Smith recommends:

In reality, what the U.S. needs to do is to A) temper our expectations for what export controls will accomplish, and B) constantly tighten the controls based on our observations of what China is doing to circumvent them.

More sanctions, lower expectations. The recommendation to do more and expect less is not a winning one. And it is eerily reminiscent of the arguments of advocates of endless sanctions everywhere. From Larison’s Eunomia on Iran:

Bobby Ghosh wants Biden to commit more fully to a policy that has already proven to be a massive failure:

“Even if Malley emerges unscathed from the investigation, the policy he has been pursuing for the president has been proven pointless. Biden should begin anew, starting with stiffer imposition of existing sanctions and the announcement of new ones as penalty for Iran’s assistance to Russia’s invasion of Ukraine.”

There are few policies that have shown themselves to be more pointless than using broad sanctions to try to compel Iran to do what the U.S. wants. Every time that the U.S. imposes new sanctions or tightens existing ones, the Iranian government has responded by doing more of the things that Washington opposes. When our government attacks other countries with economic warfare, their governments respond just as ours would if it came under attack from outside. They seek to fight back however they can, and that usually takes the form of redoubling their efforts on whatever program or policy it is that Washington demands that they abandon.

The answer with sanctions advocates always seems to be more sanctions. This is an unsustainable approach, and there is an element of un-falsifiability embedded in the logic. Any failure of the policy can be attributed to a failure of execution, the recommendation doubling down and applying more sanctions. It’s a bit like a “no true Scotsman” approach to foreign policy. (“Real sanctions have never been tried.”)

Sanctions on China’s chip industry are slightly different than broad-based economic sanctions, however, in that the destruction is the point. There is no behavioral change we are seeking in exchange for easing of sanctions. The United States has simply said, we think you should not have an advanced chip industry, because we don’t like your military modernization efforts, and then pressured allied nations to get on board with our industry-destroying-effort.

This betrays an insecurity on the part of the United States, and is in conflict with the perception of America as an upholder of important international norms and institutions which depend upon principles of fairness in competition to operate.

Perhaps the chip sanctions are necessary. I am not a scholar of foreign policy or the semiconductor industry. But they are not inevitable. And when the arguments of their supporters mimic those of the defenders of some of the most inhumane sanctions policies the United States continues to maintain, perhaps we should consider pursuing alternative courses.