Could psychology have helped Carrie Lam avoid Hong Kong’s extradition bill fiasco?

  • From confirmation bias to risk aversion, behavioural science principles shed light on Hongkongers’ reactions to the controversial extradition law

  • A better understanding of cognitive psychology could have helped the Hong Kong government

By the Hong Kong chief executive’s own admission, there were a number of shortcomings in her government’s push for the unpopular extradition bill. She has also committed to learning from these mistakes. The question is whether those shortcomings or mistakes could have been foreseen, and if anything could have been done to prevent them.

The last decade has seen an explosion of popular interest in what may be called the “behavioural sciences”, especially cognitive psychology, social psychology, and behavioural economics. Perhaps the main insight of these related disciplines is that far from being the supremely rational, self-disciplined, and interest-maximising and calculating agents that we find in standard economics, people are subject to a variety of cognitive biases and complications. Our rationality, self-control and self-interest are limited in ways that have implications for the way governments design their policies, implement their programmes, and understand decision-making.

An understanding of the behavioural sciences is critical to good policymaking and sound decision-making. Public officials operating in an environment of greater uncertainty and increasing contestation of values would do well to be familiar with the key insights of the behavioural sciences.

These include the importance of framing effects, choice architecture (or how choices are designed and presented), loss aversion (which predicts that people care much more about avoiding losses than they do about pursuing gains), prospect theory (a theory of how people assign subjective probabilities to losses and gains), and our preference for the status quo.

Indeed, it is quite possible that a better understanding of the behavioural sciences by the Hong Kong government could have prevented the fiasco that was the Fugitive Offenders and Mutual Legal Assistance in Criminal Matters Legislation (Amendment) Bill, to give the unpopular bill its official name.


While people seldom engage in the sort of rational, optimising behaviour that standard economics depicts of homo economicus, it is also incorrect to describe them or their decision-making as irrational. The cognitive biases and heuristics (or mental short-cuts) that we rely on for many of our “decisions” may not always be optimal or accurate, but they save us cognitive effort and time. They serve us well most of the time as they have been adapted through trial and error. To dismiss people’s opposition to the extradition bill as “irrational” is not just condescending, it is also psychologically incorrect.

As a general rule, unless the government has no desire to forge a consensus, it should not describe contrarian views as irrational as that delegitimises and dehumanises the opposition. This makes consensus-building far more difficult. Even if people were not fully informed about the intricacies of the bill, their concerns about the bill undermining the legal firewall between Hong Kong and the mainland were, partly at least, legitimate. Describing people who oppose your policy as “irrational” is also unlikely to be effective in persuading them to your point of view. This is due to the well-known backfire effect, or the tendency for people to reject evidence or arguments that conflict with their prior beliefs, especially if the other side pushes its point of view on them.

People also sometimes harden their positions when they are confronted with conflicting evidence; this is known as the confirmation bias. A recent psychology experiment at Stanford University asked students to respond to two studies on capital punishment: one that provided data to show that capital punishment was effective in deterring crime and the other that provided equally compelling data that capital punishment was ineffective (the studies were made up, but the students did not know that).

The students who, at the start of the experiment, had indicated that they were in favour of capital punishment found the pro-deterrence data highly credible and the anti-deterrence data unconvincing. The students who had originally opposed capital punishment reacted in the opposite way. Contrary to what rational choice theory predicts, the Stanford experiment found that those who had started out pro-capital punishment were even more in favour of it; those who had opposed it were even more opposed to it.

In the case of the extradition bill, the authorities should have anticipated that given the spate of recent, highly salient events that cast suspicion on the mainland’s justice system, it would have taken a great deal of persuasion, trust-building and time to change people’s beliefs.

What further increased people’s distrust was the undue haste with which the government tried to push through the bill. Even neutrals who may have been sympathetic to the bill initially would have found the urgency to pass the bill by June puzzling, if not suspicious.


Defenders of the bill have pointed to the fact that before the march on June 9, the government had amended the bill a second time, further limiting the scope of extraditable crimes and introducing safeguards into the agreements with jurisdictions that request extradition. So why were these changes ineffective in changing people’s minds?

Here, the work of Economics Nobel laureate Daniel Kahneman and his collaborator, the late Amos Tversky, on how people view gains and losses and assess probabilities is particularly instructive. Their Prospect Theory states the following:

People are loss averse, i.e. they value losses much more than equivalent gains;

Subjective probability weighting, i.e. people overweight small probabilities and underweight large ones;

Losses and gains are determined in relation to some reference point rather than in overall terms; and

Narrow framing, i.e. people consider risks in isolation.

Each of these propositions of Prospect Theory would have predicted that the extradition bill would be a hard sell at best, and deeply unpopular at worst.

First, it was quite clear from the outset that people cared a lot more about the losses under the proposed bill – the risks of being extradited (no matter how low these risks were), the dismantling of the firewall between Hong Kong and the mainland, the erosion of the city’s freedoms – rather than the gains in security and the criminal justice system.

This is somewhat surprising as we normally expect people to care more about their safety and security than something as abstract as their freedoms.

That the potential losses were far more salient in people’s minds in this instance was largely the consequence of the growing belief among Hongkongers, rightly or wrongly, that the mainland authorities were intent on crimping their civil liberties and protections. This raised the stakes of the bill considerably, making it unlikely that people would view the bill in dispassionate terms, or as a cost-benefit exercise comparing the gains in security against the potential loss of civil liberties. It also channelled people’s attention to what they would lose if the bill was passed, while shrouding the gains in a cloud of suspicion of the mainland’s criminal justice system. While this was unfortunate, it was also quite predictable.

Second, and crucially, the proponents of the extradition bill did not understand subjective probability weighting. They pointed out that the actual risks of Hongkongers being extradited to the mainland were extremely low; they were also confident that after the bill was passed, Hongkongers would understand the likelihood that they would be caught under the rendition arrangement was remote.

This view ignores the fact that people often exaggerate low risks and overreact to the low probability of losses by being highly risk averse. For example, people often pay too much to insure against very low risk of losses, e.g. extended warranties.

Third, losses and gains are usually not assessed holistically, but narrowly and in relation to a reference point. In this case, the reference point was the status quo. When Hong Kong was handed back to China in 1997, its lawmakers explicitly excluded the mainland from extradition and mutual legal assistance arrangements. This was accepted by the mainland government. So the “legal loophole” that the proponents of the bill referred to was not really a loophole in the sense of an omission or an oversight. The firewall between HK’s judicial system and the mainland’s is a feature, not a bug; it is there by design.

This does not mean there is never a possibility of lowering the firewall, but it does mean that any attempt to do so would require a very high bar of justification. In particular, Hongkongers will ask what has changed in the last 20 years to justify a hasty and rushed introduction of extradition/legal assistance arrangements with the mainland. The short timeline that the chief executive set for the passage of this bill was not just unrealistic, but fuelled suspicions that she was doing this at the behest of the mainland authorities. Trust is not a commodity the government has in abundance, and the way it tried to hasten the passing of the bill was hardly trust-inspiring.


The failure to anticipate the reactions of the Hong Kong people to the extradition bill by using some of the behavioural insights above was clearly a shortcoming.

Another equally important shortcoming was that the officials who sought to push through the amendments bill were also blinded by their own biases. One of these was overconfidence, which is related to, and often the result of, excessive optimism. The officials charged with the passage of the bill were unduly optimistic about the possibility of overcoming the opposition to the bill, and passing it quickly in the Legislative Council.

As Daniel Kahneman argues in Thinking Fast and Slow, “The confidence we experience as we make a judgment is not a reasoned evaluation of the probability that it is right. Confidence is a feeling, one determined mostly by the coherence of the story and by the ease with which it comes to mind, even when the evidence for the story is sparse and unreliable. The bias towards coherence favours overconfidence. An individual who expresses high confidence probably has a good story, which may or may not be true.”

n this instance, the coherent story that the chief executive may have fallen for was that her determination, toughness (in the civil service where she had served for many years, she was known as houdadak or “good fighter”), and decades of public service had made her a unique, even heroic, figure able to push through unpopular but necessary reforms in a government often criticised for inertia. This self-image may have led her to underestimate the enormous effort needed to overcome the opposition to the bill, and to overestimate her ability to do so.

Finally, it appears there was a lack of cognitive diversity among the people who worked with the chief executive on the passage of the bill. The South China Morning Post’s Jeffie Lam and Gary Cheung noted that “all three political appointees in the Security Bureau, including Justice Secretary John Ka-chiu Lee, are former police officers”; they further pointed out that the Department of Justice appeared to have provided little meaningful legal support to the Security Bureau in the first few months after the proposal was put forward in January. The bill also seemed to have involved a relatively small circle of people, even by the standards of Hong Kong’s elite policymaking process.

All this would have created a lack of cognitive diversity, which in turn meant that a piece of legislation that would have benefited from a great deal more internal scrutiny and debate never stood a chance of succeeding when it was exposed to the harsh glare of public opinion.

Donald Low is Professor of Practice in Public Policy at the Hong Kong University of Science and Technology, and Director of its Leadership and Public Policy programme. He is also the editor of Behavioural Economics and Public Policy: Examples from Singapore

External Link :