There is something deeply appealing about a plan. A problem is identified, experts examine it, evidence is gathered, a solution is designed, governments introduce legislation, institutions publish guidelines, targets are established, and everyone waits for the desired outcome. On paper, the sequence looks perfectly rational.
Then reality arrives.
People behave differently from what was expected, markets adapt, technology changes, incentives produce unexpected responses, one problem shrinks while another grows somewhere else, and the measurement chosen to demonstrate success improves while something nobody thought to measure quietly deteriorates. At that point, we have a choice. We can defend the plan, or we can learn from reality.
That distinction matters because good intentions, expertise and evidence are essential to good policy, but none of them guarantees a good outcome. A policy can be designed by intelligent people, supported by respectable institutions and introduced for entirely honourable reasons, yet still produce consequences its creators never anticipated. That is not an argument against expertise. It is an argument for humility.
Evidence itself provides a useful example. We often speak about “following the evidence” as though evidence arrives with a set of instructions attached, but evidence answers questions, and the answer depends partly on which question was asked. A study can be rigorous while examining only one part of a much larger problem. A dataset can accurately show that one indicator improved while telling us nothing about another. A policy evaluation can demonstrate that legal sales of a particular product declined without telling us what consumers subsequently purchased instead.
Sometimes the missing information matters enormously. Who wasn’t counted? What wasn’t measured? What happened outside the system being observed? What assumptions were built into the original question? Good evidence illuminates, but good judgement also notices where the light did not reach.
Risk creates a similar problem. We naturally think of reducing risk as removing something dangerous, but risk frequently moves rather than disappears. Restrict one behaviour, and another may replace it, close one market, and another may emerge; make one product prohibitively expensive, and consumers may seek cheaper substitutes; or protect people from one danger and occasionally expose them to another. Even doing nothing carries risk, yet the risks associated with the status quo often receive less attention because we are already accustomed to them.
This is why one of the most useful questions in public policy is not simply, “Is this safe?” It is, “Compared with what?” And then, “What happens next?”
Australia’s approach to tobacco and nicotine provides an extraordinary real-world example. The intention behind Australia’s tobacco policies is easy to understand: reduce smoking and nicotine use and protect young people. Those are legitimate public health objectives, but intentions aren’t outcomes.
Australia has made cigarettes extraordinarily expensive while simultaneously making regulated vaping products unusually difficult to obtain compared with many other countries. Policymakers hoped restrictions would reduce demand, but reality had other ideas. Nicotine demand did not simply disappear. Instead, enormous illicit tobacco and vaping markets developed outside the regulatory system. Criminal networks discovered that supplying products people continued to want could be extremely profitable; governments responded with greater enforcement, and illegal sellers adapted again.
Whatever position someone takes on tobacco harm reduction, it raises a much broader policy question. If you close the legal door and people simply find another entrance, what exactly have you achieved?
It isn’t enough to count what disappeared from the regulated market. We have to follow the people. What did they use instead? Where did they buy it? Was it safer or more dangerous? Was it regulated? Did overall harm decline? These are not ideological questions. They are evaluation questions.
They also illustrate something policymakers repeatedly underestimate: human behaviour. People don’t experience policy through legislation or ministerial press releases. They experience price, convenience, availability, habits, incentives and alternatives. If policymakers create a preferred option that is expensive, inconvenient or difficult to access while an undesirable alternative is cheap and readily available, telling people to choose differently will only accomplish so much.
This principle extends far beyond nicotine. Governments can prohibit something, but prohibition doesn’t automatically eliminate the desire for it. Employers can introduce procedures, but employees will find shortcuts if those procedures make ordinary work unnecessarily difficult. Parents can establish rules, but children eventually encounter situations the rules never anticipated. Technology companies can design systems expecting users to behave one way and discover millions of people using them completely differently. Human behaviour isn’t an irritating variable interfering with a perfectly designed system. It is part of the system.
Perhaps that is why I increasingly believe policy should be treated less like architecture and more like navigation. Architecture assumes we know where everything belongs before construction begins, while navigation assumes we have a destination but continuously adjust according to conditions. A good navigator doesn’t become angry at the wind because it changed direction. They adjust the sails.
Policy should work the same way. Introduce something, measure it, look for consequences beyond the original target, listen to the people affected, compare what actually happened with what was expected to happen, then adjust. Yet governments and institutions often struggle with that final step because changing policy can be portrayed as admitting failure.
I think the opposite is true. Refusing to change after circumstances change is not strength. It is stubbornness.
Imagine if medicine worked that way. A doctor prescribes a treatment expecting a particular outcome, but the patient returns six months later worse rather than better. Would we praise the doctor for saying, “The treatment was based on good evidence when I prescribed it; therefore, we must continue indefinitely”? Of course not. We would expect the doctor to reassess what happened, whether the diagnosis was correct, whether there were side effects, whether better evidence has emerged and whether another treatment is now available. We regard that process as responsible medicine. Why should public policy be different?
There is another difficulty. Once institutions invest heavily in a policy, psychologically and politically, evidence challenging it becomes uncomfortable. Careers may have been built around the approach, budgets may depend on it, organisations may have publicly defended it, and politicians may have celebrated it. Changing direction can therefore feel like losing.
This is where critical thinking becomes particularly important. The question should never be, “How do we prove our policy was right?” It should be, “What would tell us that it isn’t working as intended?”
That question changes everything. It transforms evidence from ammunition into feedback, makes unintended consequences worth discovering rather than inconvenient to acknowledge, and allows people who supported the original policy to participate in improving it without requiring them to declare everything they previously believed was wrong. Most importantly, it restores the purpose of policy itself.
Policies aren’t monuments. They are tools. Nobody keeps using a blunt tool because it was beautifully designed twenty years ago. You sharpen it, modify it or replace it. The same principle should apply to public health.
Success should never mean preserving the policy. Success should mean improving people’s lives. Sometimes that will require stronger intervention, and sometimes it will require less. Sometimes new technology will make an old approach obsolete. Sometimes an unintended consequence will become so significant that the original policy needs fundamental reconsideration, and sometimes reality will tell us that the original idea worked remarkably well. The important thing is that we remain willing to listen.
That requires a particular kind of leadership, not the leadership that claims to have predicted everything, but the leadership that can say, “We tried this for good reasons. We have learned something since then. We can do better now.” I would trust that leader far more than one who insists every previous decision was perfect.
There is something profoundly human about wanting to defend our past choices. We all do it. Changing our minds can feel like betraying our former selves, but perhaps we should see it differently. Yesterday’s decision was made with yesterday’s knowledge. Today’s responsibility is to use what we know today, and tomorrow’s responsibility will be to improve again.
That applies to tobacco policy, public health, politics, business and our own lives. The goal should never be to reach a point where we no longer have to reconsider anything. The goal should be to remain curious enough to notice when the world is teaching us something.
Because reality isn’t the enemy of a good policy. Reality is the test of it.
And when reality refuses to follow the plan, perhaps the most important question isn’t, “How do we make reality obey?”
Perhaps it is:
“What is reality trying to teach us?”
One minute. One idea. A different way of thinking.



All of us have our biases and assumptions. So it’s very important, whether doctors or policymakers, to keep checking that the medical plan is improving rather than worsening outcomes either by falling short of the objectives or else because disastrous and unacceptable side effects have developed. The following quote is often attributed to Winston Churchill although there is no record of him ever saying "However beautiful the strategy, you should occasionally look at the results". This is exactly the point Alan Gor is making. Has the policy achieved the objectives and without unacceptable adverse consequences?
Australia’s failed experiment is reflected in this statement: “Once institutions invest heavily in a policy, psychologically and politically, evidence challenging it becomes uncomfortable. Careers may have been built around the approach, budgets may depend on it, organisations may have publicly defended it, and politicians may have celebrated it. Changing direction can therefore feel like losing.”