Here is the uncomfortable part. The forty minutes are gone whether you stay or leave. So is the ticket money. Nothing you do from this point forward gets either of them back. The only thing you still control is the next eighty minutes, and the question is whether you want to spend them in that seat.

Economists call the money and time already spent a sunk cost, and the rule is simple: it should not affect what you do next. People break the rule constantly, and not because they are foolish. Leaving feels like admitting the first decision was wrong. Staying lets you avoid that, at the price of eighty more minutes.

It shows up in bigger places than cinemas. The degree you finish because you are three years in. The project a company keeps funding because of what it has already cost. The relationship that continues because of how long it has lasted. In each case the past is doing the deciding, and the past has no stake in what happens next.

A question that helps: if someone offered you this exact situation fresh, right now, with nothing already invested, would you take it? If the answer is no, the thing keeping you there is not the future. It is the receipt.

Almost nobody buys the medium. That is not a pricing mistake. The medium is not there to be bought. It is there to make the large look sensible.

Put the small next to the large on its own and $8.50 feels like a lot for popcorn. Put the medium between them and the large is suddenly only fifty cents more than something. The comparison has changed, and the comparison is what you actually use to decide. We are terrible at judging value in isolation and very good at judging it relative to whatever is sitting nearby.

This is called the decoy effect, and once you know the name you will see it every week. The subscription tier that makes no sense except as a reference point. The “most popular” plan sitting between a stripped-down one and a premium one. The house the agent shows you first, which is not for sale to you so much as for calibration.

It works on smart people because it does not ask you to believe anything false. Every number is real. It just controls what you compare against.

The defence is boring and effective. Before you look at the options, decide what you want and roughly what it should cost. Then look. The decoy only works if it gets to set the frame before you do.

Economists call this diminishing marginal utility, which is a terrible name for one of the most useful ideas you will ever learn. Each extra unit of something gives you a bit less than the one before.

Once you see it, it is everywhere. The first coffee of the day is a small miracle. The third is a habit. The first week of holiday is bliss. Week three you are checking email. A pay rise feels enormous for two months and then becomes the new normal.

Here is the practical part. Because the first unit of anything gives you the most, you get more total pleasure by spreading things out than by piling them up. One great meal a week beats three average ones. A short holiday twice a year usually beats one long one. Small, frequent, and separated wins.

It also explains why the richest people you know do not seem proportionally happier. The hundredth thousand does far less than the first.

The next time you are about to reach for the fourth slice, ask what the fourth slice is really giving you. Often the honest answer is: not the pizza.

Almost everyone would have been better off buying the passes. Almost nobody did.

The reason is not stupidity. When you sign up, you are not buying access for the person you are today. You are buying it for the person you plan to be, the one who goes three times a week. That person would get a bargain. That person also tends not to turn up.

Gyms know this. The business model quietly depends on it. A gym where every member came as often as they intended would need three times the floor space.

I do not think the lesson is “never buy a membership”. The lesson is to price things for the person you actually are. Look at the last three months, not the next three. If you went four times a month, buy the thing that is cheapest at four times a month. If that later turns out to be too little, you can upgrade, and you will be upgrading on evidence.

The version of you that goes every day is welcome to buy the membership when he shows up.

The problem is not you. It is the thousands.

In a well-known study, shoppers offered six flavours of jam bought far more often than shoppers offered twenty-four. More options should mean a better match. Instead they mean more comparisons, more chances of picking wrong, and a nagging sense that whatever you chose, something better was in there. At some point the effort of choosing exceeds the value of the choice, and people stop.

That is why you fall back on the familiar. It is not laziness. It is the only option that does not require you to evaluate the other 2,000.

Two things help. Decide what mood you are in before opening the app, in one word: funny, tense, nothing. That shrinks the set. And give yourself a rule that ends the search: three options, pick one, no going back. The rule sounds restrictive and is the opposite. It gives you back the evening.

The same thing happens with pension funds, health plans and phone contracts, with rather higher stakes than a Tuesday night. Fewer, clearer options are usually the kinder design, even though more options sounds like more freedom.

Here is the thing. He could not, and he was right not to.

Both options had roughly the same survival odds. One carried a higher chance of long-term incontinence. The other carried a higher chance of fatigue that might never fully lift. Medicine can tell you the probabilities. It cannot tell you which of those two futures you would rather live in. That is not a gap in the evidence. It is a question about your life, and the surgeon has no access to it.

What frustrated her was the handover. She heard “it’s your choice” as I am not going to help you. What he meant was closer to I have run out of things that medicine knows.

My whole field exists in that gap. When we run a choice experiment, we are asking hundreds of people to make exactly the trade her surgeon could not make for her, over and over, in slightly different combinations. From the pattern we can work out how much fatigue people will accept to avoid incontinence, on average, and how much that varies between them.

That average never tells any individual what to do. What it does is stop the system from quietly assuming. Because if nobody measures it, someone still decides. A guideline gets written, a default gets set, a funding rule gets drafted, and it embeds a trade-off that nobody ever asked a patient about.

My friend picked the fatigue. Two years on she says she would pick it again. I have no idea whether she chose correctly, and neither does she, and that is fine. What matters is that it was hers to get wrong.

You add a $14 thing you did not come for. Total: $60. You just paid $14 to avoid paying $6, and you felt good about it.

This is not a maths failure. Nearly everyone can do the subtraction. It is that the two amounts live in different mental accounts. The $6 feels like a fee, a loss, money for nothing. The $14 feels like a purchase, something you get to keep. We hate losses roughly twice as much as we enjoy equivalent gains, so a $6 loss can feel worse than a $14 spend.

Retailers set the free-shipping threshold just above the typical basket for exactly this reason. It is not a coincidence that it always seems to be a few dollars out of reach.

Two habits help. First, before adding the extra item, ask whether you would buy it at full price on its own, tomorrow, with no shipping involved. If not, you are paying for a feeling. Second, treat the shipping fee as part of the price of the things you actually wanted. A $46 basket with $6 shipping is a $52 basket. Compared like that, it is usually still the cheaper option.

None of this will stop you feeling the pull. It is meant to be felt. But you can notice it and decline.

You know these are tactics. You have known for years. They still work, and it is worth understanding why, because knowing the name of a trick and being immune to it are different things.

Scarcity does two things at once. It raises the value of the object, since things other people want feel more worth wanting. And it raises the cost of waiting, because now the option might not be there tomorrow. Both push in the same direction, and both bypass the question of whether you wanted the thing in the first place.

The tell is that the pressure comes from the seller’s timeline, not yours. You did not wake up needing to decide by midnight. Somebody chose midnight for you.

A simple habit works. When you feel the pull, ask what happens if it is gone. Most of the time the honest answer is that you will buy it somewhere else, or buy something similar, or discover you did not need it. Very occasionally it is a real loss. The point is to check, because the message is designed so that you do not.

In healthcare this one has teeth. “Limited places on the trial.” “Only a few appointments this month.” Some of it is true. Enough of it is not that the question is always worth asking.

We do not experience income as a level. We experience it as a change from a reference point, and the reference point moves. Once the higher salary becomes what you expect, it produces no feeling at all. Only a further change would.

This sounds bleak and is actually useful.

If a raise gives you eight weeks of feeling and then nothing, then the money is better spent on things that do not become reference points. Experiences fade in a different way from possessions: you stop noticing the car in your driveway within a month, but a trip you took ten years ago still produces something when you think about it. Spending on other people works similarly; it does not reset.

The other move is to keep the reference point still. If you can hold your lifestyle at the old level for a while after a rise, the gap between what you earn and what you expect to spend is where the freedom lives. It is a strange trick, feeling richer by refusing to update, but it works.

Most financial advice is about the number. Very little of it is about what the number does inside your head, which is where the whole thing is decided.

The instinct is to call this unreliability. Somebody is lying, or careless, or does not know their own mind.

I think that reading is wrong, and it took me a long time to let go of it.

Most of us have never priced a hospital wait. There is no figure filed away waiting to be reported. So when you ask, the person builds an answer, right there, out of whatever is nearby: the last bill they paid, the number you mentioned in the previous question, whether the option was on the left or the right. Change what is nearby and you change the construction.

This is uncomfortable if you want survey answers to be measurements, like temperature. It is much less uncomfortable if you treat them as what they are: a decision made under specific conditions, which you designed.

Which puts the responsibility on the person asking. If the ordering of your attributes moves the answer, then the ordering is part of your result whether you report it or not. If half your respondents never looked at the cost row, your willingness-to-pay number is partly an artefact of your layout.

None of that makes the exercise pointless. Real decisions get constructed on the spot too, in a consulting room, under fluorescent light, with someone waiting. The survey is not failing to capture a stable preference. It is capturing a constructed one, which is the only kind there is.

The useful move is to stop asking whether the number is true and start asking what it was built from.

A well-known experiment makes the point. When people were offered a premium chocolate for fifteen cents and an ordinary one for one cent, most chose the premium. When both prices dropped by a cent, so the premium was fourteen cents and the ordinary one was free, the crowd swung to the free option, even though the relative deal had not changed. Zero did something that “one cent” did not.

You can see this everywhere once you look. Free shipping tips a purchase more than an equivalent discount. A free trial converts better than a cheap one. A free gift with purchase can be worth more to a shopper than a straight price cut of the same value. The standard model says only the net cost should matter. Behaviour says the presence of a zero matters on its own.

Why? Part of it is that free removes the fear of loss. If you pay even a little and the thing turns out to be no good, you have lost something. If it is free, the downside feels like nothing, so we reach for it. Part of it is simplicity: zero needs no calculation.

This is worth knowing as a consumer, because free is often the hook on something that is not free at all, the subscription behind the trial, the data behind the app, the delivery cost folded into the price. It is also worth knowing in health and policy, where “free at the point of use” can lift take-up of a screening or a service far more than a small co-payment would predict. The lesson is not that free is a trick. It is that zero is a special number to us, and both marketers and policymakers know it.

Key reference. Shampanier, K., Mazar, N., and Ariely, D. (2007). Zero as a special price: the true value of free products. Marketing Science.

The clearest evidence comes from organ donation. Countries where people are donors unless they opt out have consent rates above ninety percent. Countries where people must opt in often sit far lower, sometimes below twenty. The populations are not that different in their values. What differs is the default. Almost no one is against donation in the opt-out countries; they simply never change the setting, just as almost no one actively signs up in the opt-in ones.

Two forces are at work. Changing a setting takes effort, and effort is enough to stop us. And a default carries a quiet signal: this is the normal, recommended choice. Together they make the preset option remarkably sticky.

This gives whoever sets the default real power, which is why it deserves care. A default can serve people, by enrolling them in a pension they would have wanted anyway, or it can exploit them, by pre-ticking an expensive add-on they would never have chosen. The same mechanism, opposite intent. The honest use is to set the default to the option most people would pick if they stopped to think, and to keep opting out genuinely easy.

For anyone designing a form, a service, or a policy, the practical point is simple: there is no neutral default. Whatever you preset will become what most people get. Choosing it well is part of the job, not an afterthought.

Key reference. Johnson, E. J., and Goldstein, D. (2003). Do defaults save lives? Science.

The trick is friction, added or removed on purpose. To keep you watching, the next episode plays automatically, so inertia carries you forward. To keep you scrolling, the feed never ends, so there is no natural stopping point. To keep you subscribed, cancelling is buried several menus deep, while signing up took one tap. Each of these leans on the same human tendencies that shape health and financial decisions: we follow defaults, we avoid effort, and we discount the future in favour of the moment.

None of this requires us to be foolish. It works precisely because these tendencies are normal. A countdown to the next episode is pleasant in the moment; the cost, another hour gone, arrives later, when the present bias that made it easy has faded.

The useful response is not guilt but design awareness. If continuing is automatic, you can add your own friction: a timer, an episode limit, autoplay switched off. If stopping is effortful, notice that the effort was put there on purpose. The same insight that helps a health service lift take-up by removing hassle can, in other hands, keep you on a platform longer than you meant to stay. Knowing which frictions were designed, and by whom, is a quietly powerful skill.

Related reading. For the underlying tendency to favour the present, see the posts on defaults and on the certainty effect; the mechanics are the same ones studied in health and policy.

The standard benchmark gives us a disciplined question: what would a decision-maker choose if their preferences were stable, comparable and internally consistent? Once we know that, behavioural evidence can ask a better question: which departures are systematic, when do they arise, and what do they imply for markets, health policy and welfare?

This is why the first step in teaching behavioural economics is not a list of biases. It is the logic of preferences, utility representation, feasible menus and budget sets. Without that foundation, behavioural economics can become a catalogue of interesting examples. With it, behavioural economics becomes a way to compare benchmark predictions with observed decision processes.

The practical implication is simple. In research and policy, do not label a choice as a bias before asking what the benchmark predicts, what information the decision-maker had, and what constraints they faced.

We read numbers left to right, and the left-most digit does more work than it should. Moving from 20,000 to 19,998 changes the leading digit from a two to a one, and our quick, intuitive sense of size latches onto that first figure before the rest catches up. The result is that a two-dollar cut can feel like a step into a lower price bracket. Retailers have known this for a long time, which is why so many prices end in 99.

The same left-digit effect shows up well beyond shopping. A treatment that lowers a risk from 10 percent to 9 percent can feel more significant than one that lowers it from 6 percent to 5 percent, although both remove one percentage point, because the first crosses from double digits to single. A patient reading 9 rather than 10 may respond differently, even though the clinical change is identical.

There is also a pull towards round numbers themselves. We set goals at 10,000 steps rather than 9,850, and we treat a round birthday as a moment for reflection more than the year before it. Round numbers feel like landmarks, so we plan and judge around them.

The practical lesson cuts both ways. As a consumer, it is worth pausing when a price sits just under a round threshold, because the gap you feel is larger than the gap that exists. And in health communication, how a number is written, single digit or double, just over or just under a round figure, can change how it lands, so it deserves the same care as the number itself.

Key reference. Thomas, M., Simon, D. H., and Kadiyali, V. (2010). The price precision effect. Marketing Science; and the broader literature on left-digit bias.

That picture is wrong in a way that matters.

When you break a mandate into its parts, who it covers, what the exemptions look like, what happens if you decline, how long it runs, people do not sort onto a line. They sort into groups that care about different parts. One group will accept almost any coverage rule but reacts badly to punitive consequences. Another has no problem with consequences but wants medical exemptions handled by a doctor rather than a form. A third mostly wants to know when it ends.

That has a practical consequence I did not anticipate. Two mandates with the same overall support can be supported by completely different people. Soften the penalty and you gain one group while losing another, and the headline number barely moves. If you are only watching the headline number, you conclude the change did not matter. It mattered enormously to who was in the room.

It also explains something that puzzled me early on. Public opposition to mandates often spiked over details that looked administratively trivial to the people designing them. The exemption paperwork. The review date. Those were not trivial to the groups they landed on.

I now think the useful question is rarely “will the public accept a mandate”. It is “which mandate, and who does each version lose”. The first question has a number for an answer. The second has a design.

Three points help when you read a DCE.

The first is that a DCE measures stated preferences, not observed behaviour. People tell us what they would choose in a described situation. This is a strength when we want to study options that do not exist yet, such as a new vaccine or a service that has not been funded. It is also a limitation, because what people say and what they do can differ. A good study is designed to make the choices realistic and to narrow that gap, but the gap does not vanish.

The second is that the useful result is usually a trade-off, not a single number. The interesting finding is rarely that people like a higher success rate. It is how much waiting time, or how much cost, people will accept for a given gain in success. Trade-offs are what decisions are made of, so they are what a DCE is built to estimate.

The third is that the average can mislead on its own. A DCE gives an average preference across the sample, but people differ, sometimes a lot. Two groups can produce the same average while wanting opposite things. I will come back to this in a separate post, because it is often the most important part of a study for policy.

So when you read a DCE, look for what was on offer, which is the attributes and their levels, then what people traded off, and finally whether the study looked beyond the average. Those three things tell you most of what you need to know.

The benchmark struggles here. A single risk attitude, captured by the curvature of a utility function, cannot easily produce a taste for a long-shot gain and a taste for protection against a long-shot loss in the same person. Something other than the size of the outcomes is doing the work.

That something is how we treat probabilities. A robust finding is that people tend to overweight small probabilities and underweight moderate to large ones. A one-in-a-million jackpot feels more likely than one in a million; a small chance of a house fire also feels more likely than the raw number suggests. Overweighting the small chance of a large gain makes the lottery ticket feel worth it. Overweighting the small chance of a large loss makes the insurance feel worth it. The two behaviours are two faces of the same weighting.

A quick illustration. Suppose a lottery pays A$2,000 with probability 0.01, and the alternative is A$30 for certain. On average the lottery is worth A$20, less than the sure A$30. But if that one percent chance is treated as if it were five percent, the lottery can feel worth about A$100 in decision terms, and it wins. Nothing about the payoffs changed; only the weight on the probability did.

This is a descriptive account of choices, not an endorsement of either purchase. Insurance can be entirely sensible when a loss would be ruinous, and buying the occasional ticket for entertainment is a personal matter. The behavioural point is narrower: a model that assumes probabilities enter linearly will misread both markets, and consumer protection that focuses only on the true odds may miss how the odds are perceived.

For policy this matters in two directions. Where small risks are overweighted, people may over-insure against minor losses or be drawn to lottery-like products that are poor value. Where larger risks are underweighted, people may under-prepare for events that are quite likely over a long horizon. Good communication works with this, presenting probabilities in ways that are easier to gauge rather than assuming the raw number lands accurately.

The certainty effect is the near-certainty end of the same weighting function, and the ECON3111 resources work through the arithmetic.

Key references. Kahneman, D. and Tversky, A. (1979). Prospect theory: an analysis of decision under risk. Econometrica. Tversky, A. and Kahneman, A. (1992). Advances in prospect theory: cumulative representation of uncertainty. Journal of Risk and Uncertainty.

Preferences differ across people. In health they often differ in ways that line up with who is affected by a decision. Older and younger people, people with and without a condition, and people in cities and in rural areas can weigh the same features very differently. When we collapse all of that into one number, we can end up describing a person who does not exist: the average of two groups who want opposite things.

This matters for policy in a direct way. If a programme is designed around the average preference, it may fit no one well. It may also work against the people a policy is meant to help, if their preferences sit far from the middle. The question of who a policy is for, and how their preferences differ from everyone else’s, is not a technical footnote. It is often the decision.

In choice modelling we have tools for this. Mixed logit models let preference weights vary across people rather than fixing them at a single value. We can look at how preferences relate to characteristics we can observe, and we can describe the spread, not just the centre. None of this is exotic, and it changes what a study can say.

My advice when reading any preference study is simple. Do not stop at the headline average. Ask whether the study looked at variation, who differs from the mean, and by how much. If the answer is only an average, treat it as a first sentence, not the whole story.

The benchmark evaluates outcomes by their final level. In the standard model a dollar is a dollar, and how you arrived at it does not matter. Prospect theory changes the reference point. Outcomes are judged as gains or losses relative to where you started or what you expected, and losses are weighted more heavily than equal-sized gains. A common working figure is that a loss feels roughly twice as significant as a gain of the same amount, though the exact number varies.

This has direct consequences. Consider warranties. A repair bill is felt as a loss relative to the reference point of “no repair cost,” and the pain of that loss makes paying a premium to avoid it more attractive than the expected cost alone would justify. The same logic helps explain why out-of-pocket health payments can feel heavier than a premium of the same size that is bundled into a regular deduction: one is coded as a loss at the moment of care, the other is a smaller, expected background cost.

Framing follows from the same idea. “Nine out of ten patients keep their mobility” and “one in ten patients lose mobility” describe the same fact, but the loss frame lands harder. This is not a trick to be deployed casually; in health communication it is an ethical matter, because the frame that feels most persuasive is not always the one that helps a person decide well. The honest use of this knowledge is to present both frames, or to choose the frame that supports understanding rather than the one that maximises a desired response.

It is worth separating description from prescription. Loss aversion describes how outcomes are experienced; it does not say that treating a loss as worse than a gain is a mistake. In many settings caution about losses is reasonable. The behavioural claim is that the asymmetry is systematic and predictable, so a model that ignores reference points will misjudge choices about prices, insurance, and treatment.

The limits matter too. Reference points can shift, expectations adapt, and the size of loss aversion depends on the domain and the stakes. It is a strong regularity, not a universal constant.

For the underlying model, see the ECON3111 resources; for the way framing interacts with public-health messages, the work on preference research and vaccine policy is a useful companion.

Key references. Kahneman, D. and Tversky, A. (1979). Prospect theory: an analysis of decision under risk. Econometrica. Tversky, A. and Kahneman, A. (1991). Loss aversion in riskless choice: a reference-dependent model. Quarterly Journal of Economics.