Crisps Out, Yogurt In: What Ozempic Reveals About Consumer Choice

It is a Wednesday evening in an English supermarket. The snack aisle, usually the busiest corner of the shop, is strangely untouched: bags of salted chips remain stacked, and the biscuits have barely moved. One aisle over, the Greek yogurt fridge is half empty, with each trolley having a different brand. Nothing about the prices has changed. There is no discount, no new health study, no sudden drop in income. And yet, across millions of households, shopping wish lists are subtly being rewritten. The explanation lies not in the store, but in a weekly injection. Economics has always assumed that our tastes are our own. The yogurt aisle suggests they might be available at the pharmacy now.

1. Brief background on GLP-1s

GLP-1 drugs were not developed to change anyone’s shopping habits. They mimic glucagon-like peptide-1, a hormone the gut naturally releases after a meal to signal satiety. The drugs strengthen and prolong that signal: they trigger insulin release from the pancreas, reduce glucose production in the liver, and make users feel full while slowing how quickly they feel hungry again. Many of these users describe the constant “food noise” in their consciousness as simply having gone quiet, according to a survey done by the European Association for the Study of Diabetes.

First sold as Ozempic to treat blood sugar regulation, the same molecule now appears as Wegovy, used for weight loss, alongside rival injections such as Mounjaro and Zepbound, though their reach is no longer marginal. According to Gallup, 11% of American adults now take GLP-1 drugs for weight loss, up from 3% in 2024, and the new Wegovy pill passed two million prescriptions within months of its January launch.

2. Crips Out, Yogurt In

If craving really changes, it should show up somewhere measurable, and the most honest documentation of what people want is what they buy. In late 2025, three Cornell economists, Sylvia Hristakeva, Jūra Liaukonytė and Leo Feler, set out to find it. They connected survey answers about medication use to the transaction data of 150,000 American households, tracking what ended up in shopping trollies before and after someone started a GLP-1 drug.

Within six months of starting the medication, households cut their grocery spending by an average of 5.3%, and spending at fast-food chains and coffee shops fell by about 8%. The reduction was far from even. Savoury snacks suffered most, falling 10.1%, while yogurt was the only category with a statistically significant increase. Wealthier households modified their spending the most, cutting grocery spending by 8.2%.

One limitation is that the data cannot fully separate the drug’s biological effect from other lifestyle changes users might make at the same time. Still, the pattern is hard to ignore. Companies have noticed. Danone, maker of Oikos Greek yogurt, reports double-digit growth in its high-protein products as GLP-1 use spreads, while food company shares fell when the Wegovy pill was approved. Investors, it seems, expect the change to last.

Crisps lose, Greek yogurt wins, and nothing about the pricing of either has changed.

3. There’s No Debate About Taste. Until Now.

To understand why this matters, it helps to know how economists usually explain a change in purchases. In the standard model, demand depends on three factors: price, income, and preferences. If people buy fewer crisps, the textbook suspects that crisps became more expensive or that people became poorer. Preferences, by contrast, are treated as fixed.

This was not an accident but a deliberate choice. In 1997, George Stigler and Gary Becker, both later Nobel laureates, published a paper whose title says it all: De Gustibus Non Est Disputandum (there is no arguing about taste). Their argument was that economists should assume tastes neither change capriciously nor differ majorly between people. They argued that justifying behavior by saying “tastes changed” is too easy, it can explain anything, and therefore explains nothing.

Economics also has a way of analyzing those tastes. The theory of revealed preference, developed by Paul Samuelson in 1938, holds that we do not need to ask people what they want, as we can simply watch what they choose. Your shopping basket, in other words, is your visible preference.

GLP-1 drugs create a critical situation for both ideas. In the Cornell data, nothing changed in prices or income, yet trolleys did. A change in taste was triggered by a mass prescription.

4. The Planner, the Doer, and the Drug

If the story ended here, it would simply be a case of tastes changing. But the Cornell data contains a twist. Around a third of users stopped taking the medication during the study, and their food spending returned to its pre-drug level, with trolleys even slightly less healthy than before. The new preferences did not last. They had been switched on, and they could be switched off.

On the drug, a shopper buys yogurt; however, off of it, they go back to crisps. If our purchases reveal what we really want, which one is the real choice?

Behavioral economics offers a way to reason this. In 1981, Richard Thaler and Hersh Shefrin proposed the planner-doer model, in which each of us contains two selves. The planner thinks about the long term (health, savings, next year), while the doer lives in the moment and wants the crisps now. Much of daily life, in this life, is the planner trying to keep the doer under control.

To win that fight, planners use commitment devices, which are tools that restrict their future options on purpose. Ulysses had himself tied to the mast so he could hear the Sirens without steering towards them. Today, people delete shopping apps, set up automatic savings, or keep no biscuits in the house. Seen this way, a GLP-1 drug looks less like a diet but more like a commitment device. It does not hide the crisps, but quiets the doer who wants them.

This leaves economists in an unusual position. Traditionally, economics does not judge what people want. If someone prefers crisps to yogurt, that preference is simply taken as given, and the economist’s job is to explain how markets serve it. A person is considered better off when they get more of what they want, whatever that happens to be.

Because the doer sometimes overrules the planner, Richard Thaler and Cass Sunstein argued in Nudge (2008) that gentle interventions, like placing fruit at eye level, can help people make the choices they would make “if they were thinking clearly”. This picks the side that the planner’s preferences are treated as the real ones, and the doer’s as mistakes.

However, when it comes to the GLP-1 drug, it changes the person who is choosing. On it, the shopper is genuinely content with yogurt, but off it, they are genuinely content with crisps. When choices flip like this, economists Douglas Bernheim and Antonio Rangel (2009) argued that there may be no single “true” preference to measure at all.

So, the question “what does the consumer want?” may no longer have one answer. It may depend on what they took this morning.

5. Preferences on Prescription

For most of its history, economics has emphasized that the consumer is sovereign. Our choices steer markets, firms compete to satisfy them, and our preferences, no matter what, are taken as our own. GLP-1 drugs complicate that picture. If a weekly injection, or now a daily pill, can quiet one set of desires and amplify another, then preferences are no longer just a starting point for economic analysis. They can now be prescribed.

Does that make economics outdated? Not quite, and for three reasons.

First, the discipline has argued about changing tastes for decades. In 1958, John Kenneth Galbraith argued in “The Affluent Society” that advertising creates the very wants that production then satisfies, a “dependence effect” that undermines consumer sovereignty. Even Gary Becker wrote “Accounting for Tastes”(1996), modeling how habits and social pressure shape what we want. Ozempic does not reveal a blind spot nobody had seen; it just makes an old dispute hard to ignore.

Second, the drug actually confirms much of what economics predicts. When demand for snacks fell and demand for yogurt rose, firms such as Danone responded by expanding its high-protein range, and investors marked down the value of food companies as soon as the Wegovy pill got approved. The tools for explaining what happens after preferences change still work remarkably well. What economics struggles with is explaining the root of the change itself.

Third, assumptions in economics are tools, not beliefs. Milton Friedman argued in 1953 that a model should be judged by how well it predicts, not by how realistic its assumptions are. Stable preferences were always a simplification, so it is just a matter of whether that simplification still serves us.

Is the shopper better off on the drug because they now want yogurt, or off it because they once again want crisps? Daniel Kahneman offered one way out, distinguishing between the “decision utility” revealed by our choices and the “experienced utility” of how we actually feel. Asking how people feel, not what they buy, may become essential.

All in all, economics is not outdated. But one of its most subtle assumptions has met its first real-world test, and the discipline will have to answer a question it long preferred to avoid, which is not just what people want to buy, but also who decides what they want.

References:

Arnaut, T. et al. (2025) ‘Impact of food noise after initiating semaglutide treatment: results from a US survey (INFORM)’, 61st Annual Meeting of the European Association for the Study of Diabetes, Vienna, 15–19 September.

Becker, G.S. (1996) Accounting for Tastes. Cambridge, MA: Harvard University Press.

Becker’s Hospital Review (2026) GLP-1 use hits record high in US as obesity rates fall: 6 things to know. Available at: https://www.beckershospitalreview.com/glp-1s/glp-1-use-hits-record-high-in-us-as-obesity-rates-fall-6-things-to-know/(Accessed: 19 October 2026).

Bernheim, B.D. and Rangel, A. (2009) ‘Beyond revealed preference: choice-theoretic foundations for behavioral welfare economics’, Quarterly Journal of Economics, 124(1), pp. 51–104.

Fox 29 (2026) Here’s how many Americans say they are now taking GLP-1 drugs, 10 July. Available at: https://www.fox29.com/news/heres-how-many-americans-say-now-taking-glp-1-drugs (Accessed: 19 October 2026).

Friedman, M. (1953) ‘The methodology of positive economics’, in Essays in Positive Economics. Chicago: University of Chicago Press, pp. 3–43.

Galbraith, J.K. (1958) The Affluent Society. Boston: Houghton Mifflin.

Hristakeva, S., Liaukonytė, J. and Feler, L. (2024) ‘The No-Hunger Games: how GLP-1 medication adoption is changing consumer food demand’, SSRN Working Paper. Available at: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5073929 (Accessed: 19 October 2026).

Hutt, W.H. (1936) Economists and the Public. London: Jonathan Cape.

Kahneman, D., Wakker, P.P. and Sarin, R. (1997) ‘Back to Bentham? Explorations of experienced utility’, Quarterly Journal of Economics, 112(2), pp. 375–405.

McAdams, R.H. (1997) ‘Comment: accounting for norms’, Wisconsin Law Review, 1997, pp. 625–637. Available at: https://chicagounbound.uchicago.edu/cgi/viewcontent.cgi?article=1033&context=journal_articles (Accessed: 19 October 2026).

Reiley, L. (2025) ‘Ozempic is changing the foods Americans buy’, Cornell Chronicle, 19 December. Available at: https://news.cornell.edu/stories/2025/12/ozempic-changing-foods-americans-buy (Accessed: 19 October 2026).

Reuters (2025) ‘Weight-loss pill shakes food industry’, The Western Producer. Available at: https://www.producer.com/farm-family/weight-loss-pill-shakes-food-industry/ (Accessed: 19 October 2026).

Samuelson, P.A. (1938) ‘A note on the pure theory of consumer’s behaviour’, Economica, 5(17), pp. 61–71.

Smith, A. (1776) An Inquiry into the Nature and Causes of the Wealth of Nations. London: W. Strahan and T. Cadell.

Stigler, G.J. and Becker, G.S. (1977) ‘De gustibus non est disputandum’, American Economic Review, 67(2), pp. 76–90.

Thaler, R.H. and Shefrin, H.M. (1981) ‘An economic theory of self-control’, Journal of Political Economy, 89(2), pp. 392–406.

Thaler, R.H. and Sunstein, C.R. (2008) Nudge: Improving Decisions About Health, Wealth, and Happiness. New Haven: Yale University Press.

Follow our socials

Community

© 2026. All rights reserved.