Probably not. People are likely to eat smaller and fewer meals, but not necessarily healthier ones.
Wall Street has placed a big bet on GLP-1 drugs. When Walmart's U.S. CEO, John Furner, told Bloomberg in October 2023 that customers on Ozempic were buying "slightly less calories," packaged-food stocks cratered. The five largest pure-play companies fell roughly 18% while the S&P 500 gained 24%. Prices stabilized in 2024 when analysts concluded the GLP-1 fears were overblown. Then they fell again in 2025. General Mills, Campbell's, and Conagra are now down more than 50% from their peaks.
Goldman Sachs has projected these drugs could add more than 1% to gross domestic product if adoption reaches 60 million Americans by 2028, driven by higher labor-force participation and lower absenteeism among healthier workers. The investment thesis is baked into valuations across food, retail and healthcare, and Wall Street shows no sign of second-guessing it.
Such excitement may be warranted on the health front. In clinical trials, semaglutide (Ozempic and Wegovy) produced average body-weight reductions of 14.9% vs. 2.4% on placebo; tirzepatide (Zepbound and Mounjaro) achieved up to 20.9% at the highest dose. A landmark cardiovascular trial found that semaglutide reduced the risk of heart attack, stroke and cardiovascular death by 20% in high-risk patients. The FDA has approved the drug for kidney disease and sleep apnea, and emerging evidence points to benefits in liver disease, arthritis and addiction.
But the leap from clinical trials to claims that these drugs are transforming how America shops and eats rests on evidence so thin that any first-year economics student should be skeptical. The most-cited consumer research compares GLP-1 users with nonusers and finds striking differences. A 2024 KPMG analysis estimated that GLP-1 users consumed 21% fewer calories. Cornell University researchers, publishing in the Journal of Marketing Research in December 2025, found that households with at least one GLP-1 user reduced grocery spending by 5.3% and spending at fast-food and coffee chains by 8% within six months of starting the drug.
Every one of these sources has the same underlying problem: The people taking GLP-1 drugs aren't a random cross-section of Americans. According to Gallup, 12.4% of U.S. adults currently report using GLP-1 for weight loss -- a figure that more than doubled in roughly 18 months. GLP-1 users skew heavily toward the affluent and health-motivated; they are more than twice as likely to have household incomes exceeding $100,000 annually as the general population. The kale-buying, snack-aisle-skipping behavior observed in current users most likely reflects their own inclinations, not merely what the drug does.
Consider yoga pants. When Lululemon launched in the early 2000s, its customers were athletic, health-conscious and already eating well. If you had surveyed Lululemon buyers in 2004 and compared them with nonbuyers, you would have found striking differences in diet, exercise and grocery spending. The temptation would have been to conclude that yoga pants were transforming their relationships with their bodies. That would have been wrong. You were measuring the Lululemon customer, not the pants. When yoga pants reached Walmart, those behavioral differences disappeared.
GLP-1 drugs unambiguously suppress appetite. Evidence suggests the drugs reduce what clinicians call "food noise": intrusive, reward-driven cravings for high-calorie foods. Small trials in laboratory settings show that semaglutide and tirzepatide reduce the motivation to eat high-fat savory foods. Other studies show reduced activation in reward-related regions of the brain in response to food cues. But whether all these lab-based results translate into real-world dietary changes has never been tested in any large randomized trial.
None of the seven landmark GLP-1 trials measured changes in dietary composition, food quality or food choice. We also don't know how many people will stay on these drugs.
Real-world data suggests roughly half of patients prescribed GLP-1 for weight loss discontinue within 12 months. What fraction will become lifetime users is unknown. The post-discontinuation news, too, is sobering: Weight regain begins quickly, and a 2026 study published in BMJ Medicine by Yan Xie, Taeyoung Choi and Ziyad Al-Aly found that interruption increases the risk of heart attack, stroke and death by 14% and 22%, respectively, for those off the medication for one and two years, although patients who stop may differ from those who persist in ways that independently predict worse outcomes.
All this matters for the fiscal math underlying optimistic GLP-1 projections. The healthcare savings thesis -- reduced hospitalizations, lower disability rates, improved labor productivity -- is built on clinical trial populations. It assumes health benefits that compound over decades of sustained use. A world in which half of users cycle on and off the medication is a world in which those benefits are partial and intermittent, the costs are recurring and the long-run dividend could be a fraction of what models project.
As GLP-1 adoption broadens, moving down the income distribution to patients who are less health-motivated to begin with, dietary-choice effects in the data will fade toward zero. The reduction in calories consumed may persist -- because that reflects pharmacology -- but the effects of GLP-1 users eating fresh produce and avoiding processed food will diminish, because those selections always reflected the early customers, not the drug. If late adopters eat half a Big Mac on Ozempic rather than switching to a salad, we will know the healthy-eating signal was a product of self-selection all along. It will be clear that food preference shifts observed in lab settings weren't strong enough to override real-world habits across the general population.
This prediction can be evaluated once Medicare and Medicaid coverage expands. If the early effects hold across income groups, the consumer-transformation thesis gains credibility. If they fade, the food industry pivots under way will look like a very expensive bet on a small initial sample.
The drugs are effective. The revolution in how Americans shop and eat is, for now, a story we are telling about wealthy, health-motivated Americans -- people willing to pay out of pocket to inject themselves weekly to lose weight. When packaged food stocks crashed a second time, in 2025, some of the repricing was warranted: The evidence that GLP-1 users buy less had hardened. But buying less is pharmacology. The multibillion-dollar pivot to GLP-1-friendly products still rests on the same contaminated evidence it always did. Half a Big Mac isn't a salad.
Mr. Fryer, a Journal contributor, is a professor of economics at Harvard, a founder of Equal Opportunity Ventures and a senior fellow at the Manhattan Institute.