Published by The Lawfare Institute
in Cooperation With
On June 4, the Supreme Court of New Zealand issued an important judgment in a case involving the extradition to China of Kyung Yup Kim, a South Korean national resident in New Zealand wanted on a murder charge. Initial reports about the 150-page judgment—that it was a 3-2 vote clearing the way for Kim’s extradition—are not exactly wrong, but they convey a misimpression as to both the split of opinion and the effect of the judgment. As it stands, the government is not yet permitted to extradite Kim. At the same time, however, all five judges agreed that it could do so if the Chinese government provided appropriate assurances that Kim would receive a fair trial and not be tortured.
While liberal democratic countries such as Canada and the United States have on occasion delivered wanted suspects to China, whether under the rubric of extradition or some other proceeding such as deportation, the cases have involved suspects who were, at least originally, Chinese citizens and not citizens of the countries where they had sought refuge. This case is relatively rare in involving a suspect who has never been a Chinese national. (In 2019, however, Spain extradited to China 94 Taiwanese nationals wanted on telecom fraud charges despite Taiwanese government protests. And in previous years, Taiwanese nationals have been extradited to China by Armenia, Cambodia, Kenya, Malaysia, and the Philippines.)
It could set a troubling precedent not just in New Zealand but also in other liberal democratic countries whose judiciaries may be impressed by their sister court’s unquestionably thorough examination of the Chinese judicial system and the issues at stake. Unfortunately, the decision rests heavily on a plausible but demonstrably flawed premise: that China’s concern for reputational damage will ensure that it keeps its promises. The record shows that China is willing to violate its international commitments in criminal justice matters when it finds it convenient, and granting extradition in this case risks opening the door to further extraditions on the basis of unreliable guarantees.
The Procedural Posture
The Supreme Court’s judgment is not the final ruling in the case. In the New Zealand system, the government—in the person of the minister of justice—makes a determination to extradite (a “surrender decision”), and the defendant may then seek review of that decision in a court. There is no extradition treaty between China and New Zealand, but extradition is nevertheless possible if certain legal standards are met. The defendant argues, of course, that those standards are not met.
The first surrender decision was made in November 2015. On review, the High Court determined that various assurances received from China were inadequate and sent the surrender order back for reconsideration. On reconsideration, the minister again decided to extradite. That decision was upheld by the High Court, which was satisfied that the deficiencies had been addressed. The Court of Appeal, however, reversed the High Court, identifying “wide-ranging” concerns with the minister’s decision. It was this Court of Appeal judgment that was before the Supreme Court.
What the Supreme Court did was neither to affirm nor to reverse the Court of Appeal. Instead, much like the High Court in the first surrender decision and the Court of Appeal in the second surrender decision, it identified issues regarding which it wanted further and clearer assurances, and gave the government until July 30 to produce them—technically, adjourning the appeal. The only issue on which the court split 3-2 was whether this kind of adjournment, which gives the government another bite at the apple within the same proceedings, was appropriate.
On other key matters, however, the judges were unanimous. First, they unanimously shared the view that the assurances received so far were inadequate. The minority differed from the majority only in believing that the necessary consequence of that view was upholding the Court of Appeal and sending the matter back to the minister for yet more reconsideration. (The majority seemed moved largely by the desire to prevent further delay in the proceedings.)
But—and this is critical—all judges shared the view that extradition to China was in principle permissible; there were no general grounds on which to rule it out. All shared the view that given appropriate and credible assurances, this suspect—and presumably any suspect—could be extradited.
To sum up, the judgment says that Kim’s extradition is permitted if the government can, by July 30, procure from China further and better assurances than those it already has.
The decision dealt with two main questions. If extradited to China, would Kim (a) be tortured, and (b) receive a fair trial? Each question then generated a host of subsidiary questions: Mainly, what counts as “torture” and “fair,” the standards of probability and of proof, and what kinds of assurances, if any, could be accepted as mitigating the relevant risks.
Whatever one thinks of the decision, it cannot be said that the judges and lawyers did not do their homework. The text and footnotes in the judgment reveal a thorough and wide-ranging inquiry into the Chinese legal system, including testimony and scholarly works that paint a far from rosy picture. This is the most thorough examination of the Chinese legal system that I have seen from any court in any country. (I have not, of course, seen them all, but I am fairly confident that no U.S. court decision matches this in its thoroughness.)
On the torture issue, the court—bound by the standards of the U.N. Convention Against Torture—found the assurances adequate and held for the government. I interpret the minority as concurring in this substantive part of the judgment.
The fair trial issue is more important and more complex. The court’s analysis of whether Kim could receive a fair trial in China is long and detailed. Among other things, the court was particularly concerned about the possibility that the case would be submitted for final decision to a Chinese court’s adjudication committee, a group of senior court officials who would not have attended the trial and that includes an official from the prosecutor’s office as a nonvoting member. The court said in effect that if adjudication committees operated in the way they usually operate, that would be a per se bar against extradition—yet the court seemed determined to avoid making a per se ruling. It tried to recharacterize adjudication committee proceedings in a way that made them seem acceptable, suggesting at one point that they might be “akin to a preliminary general appeal decided on the papers.” And it suggested that suitable assurances about adjudication committee proceedings in Kim’s particular case could solve the problems, even though—and this it did not acknowledge—whether such assurances were faithfully implemented could not possibly be monitored.
A key issue is whether the desired assurances, if offered, can be trusted. The court generally took the view that China could be trusted to keep its promises, and generally not to mistreat Kim, because it would be concerned about reputational damage should it be perceived as a promise-breaker. It cited Fu Hualing, a respected scholar and dean of the University of Hong Kong’s Faculty of Law, to the effect that China “desperately” needs international cooperation in criminal matters and so would be unwilling to incur the reputational cost of (for example) torturing Kim to confess.
All of this seems to make a lot of sense and is intuitively appealing. But it can be tested empirically, and has been. It turns out not to hold water.
In a number of cases, the Chinese government has clearly abused the legal process for political purposes—the continuing detentions of Canadians Michael Kovrig and Michael Spavor as hostages for the release of Meng Wanzhou, or the exit ban on Americans Victor and Cynthia Liu in order to force the return of their father, who is wanted on corruption charges, are just some recent examples. Fear of reputational damage has not constrained the government in these cases. China indeed wants, and perhaps even desperately needs, international cooperation in criminal matters. But it does not yet seem willing to pay for it in the coin of due process.
On the specific issue of assurances, there are cases in which China has violated clear treaty commitments, again apparently unconcerned about reputational damage. That lack of concern, unfortunately, seems to be justified; governments continue to enter into treaties with China and put faith in the government’s promises. The court’s willingness in principle to do so here is a case in point.
These violations are not matters of subjective judgment or matters of degree. They are clear-cut. China’s commitments under its consular relations agreements with Australia and Canada to allow diplomatic representatives to attend trials of their nationals contain no exceptions, and the agreements specifically state that China may not invoke its own domestic law as a reason to deny access. Yet China has done this on multiple occasions: To take just three examples, at the trial of Australian Stern Hu in 2010, at the trial of Canadians Michael Kovrig and Michael Spavor in March 2021, and at the trial of Australian Yang Hengjun last month. It has also refused to respect its promises of consular access to Canadian Huseyin Celil and Swede Gui Minhai, arguing—contrary to the provisions of China’s own Nationality Law—that the former never lost his Chinese citizenship and that the latter voluntarily renounced his Swedish citizenship.
The court notes that assurances of the type sought are not as binding as formal treaty commitments. Yet China has openly and clearly violated its treaty commitments on matters of criminal justice. How much less reliable, then, are these concededly less binding assurances? Despite the intuitive appeal of the argument about reputational damage, history simply fails to support it. In the past, China has ignored its promises when convenient and asserted its “judicial sovereignty”—as if that somehow meant that promises did not have to be kept. There is no reason to imagine that the future will be different from the past.
A Precedent, Not an Exception
This is an important test case for China. It is apparently the first time China has requested extradition from New Zealand, and it chose a suspect against whom the evidence, at least as it appears in the court’s judgment, is pretty strong. (In its discussion on torture, the court discounted its likelihood precisely because it believed that a confession would not be necessary to establish guilt, the other evidence being sufficient.) And the suspect’s apparent guilt may have led the court to convince itself that a fair trial could take place, or that China’s assurances could be believed. But once this case becomes a precedent, nobody is going to do the work of looking up all the details—at least, nobody except the defense lawyers, but by then the Overton window on extradition will have shifted and a defense case that might have been easy before this case will have become more difficult.
Editor's note: This piece has been corrected to qualify the claim that it is "rare" for a non-Chinese national to be extradited and has been updated with other examples of this happening.