Gary Taubes on Why We Get Fat

Episode Summary

Executive Summary: Russ Roberts and Gary Taubes debate the causes of obesity, challenging the simple “eat less, exercise more” model. Taubes argues obesity is primarily a hormonal/metabolic disorder driven by refined carbohydrates and insulin, not just excess calories or inactivity, and emphasizes that behavior like hunger may be a result of fat-tissue dysregulation rather than the cause.

Main Topics: Critique of Calories In, Calories Out (Priority: 5/5): Taubes rejects the idea that obesity is best explained as a simple energy-balance problem, arguing that it ignores the body’s internal hormonal regulation and reduces humans to black-box thermodynamics. Insulin and Carbohydrate Hypothesis (Priority: 5/5): The central alternative model is that insulin regulates fat storage, and refined carbohydrates raise insulin, promoting fat accumulation even when total calories are not high. Historical Obesity Research and Forgotten European Science (Priority: 4/5): Taubes contends that pre-World War II European researchers, especially German and Austrian scientists, understood obesity as a hormonal regulatory problem, but their work was largely ignored in the U.S. after the war. Behavior as an Effect, Not a Cause (Priority: 5/5): The conversation explores how hunger, cravings, and overeating may result from physiological changes in fat tissue rather than simply causing weight gain, using examples like growing children, tumors, and rat experiments. Evidence from Populations and Extreme Cases (Priority: 4/5): Taubes cites historical and cross-cultural examples of obesity in poor or physically active populations to argue that inactivity and abundance alone cannot explain weight gain. Exercise, Dieting, and Weight Loss Limits (Priority: 4/5): Roberts and Taubes discuss why exercise and calorie restriction often fail long-term: they may reduce carbs somewhat, but they also increase hunger and are hard to sustain, especially if the underlying hormonal drivers remain unchanged. Genetics, Individual Variation, and Self-Experimentation (Priority: 3/5): They consider genetic predisposition, twin examples, and personal dietary experiments, concluding that carbohydrate sensitivity varies but that genetics likely interacts with diet rather than fully determining obesity.

Key Arguments: Obesity should be understood as a disorder of fat-tissue regulation, not merely a matter of overeating and inactivity. Insulin is a central hormone in fat storage; carbohydrates raise insulin and can drive fat gain independent of total calories. The body’s hunger and cravings may be downstream effects of fat-tissue dysregulation, not simply signs of weak willpower. Historical and modern examples of obesity among poor, hardworking, or physically active people contradict a simple “lazy + overeating” explanation. Low-carb, higher-fat eating can allow some people to lose weight with less hunger because it changes the hormonal environment governing fat storage. Exercise alone does not reliably prevent obesity; many highly active people remain overweight, and elite athletes may be a selected, nonrepresentative sample. Twin and animal examples suggest biology and hormones can create obesity phenotypes even under similar environments, implying strong individual susceptibility. Nutrition science is difficult because human trials are hard to design cleanly; Taubes argues better experiments are needed before concluding the calorie model is correct.

Data Points: Date of episode: July 11, 2012 - Opening metadata for the EconTalk conversation. Prior appearance on EconTalk: November 2011 - Roberts notes Taubes was previously on the show to discuss Good Calories, Bad Calories. Obesity epidemic timeframe: About 40-50 years - Taubes notes obesity and diabetes rates rose dramatically without a corresponding genetic change. Pima Indians famine period: 20 years - Russell observed obesity among Pima after decades of famine in the late 19th century. Pima observation year: 1902-1906 - Historical example of obesity documented by anthropologists in the early 20th century. Trinidad diet estimate: No more than 2,000 calories/day - MIT nutritionists studying Trinidad noted obesity despite limited caloric intake. Trinidad fat composition: 21% fat - Reported composition of the Trinidad diet in that example. Running/weight example: 5-6 marathons a year - Taubes cites an endurance-running expert who gained 40-50 pounds despite frequent marathons. Endurance-athlete example: 90-minute weight workout + 70-mile bicycle race - Taubes describes a colleague adapted to ketosis who can do major exercise without eating until dinner next day. Body fat example: 6% body fat; about 9 pounds of fat on a 160-pound person - Used to illustrate that even very lean bodies have substantial energy reserves. Traditional calorie deficit figure: 3,500 calories per pound of fat - Roberts references the common rule-of-thumb underlying calorie counting. Insurance/response rate analogy: Only about 10% of smokers get lung cancer - Taubes uses smoking as an analogy for risk not being universal even when causation is real. Smoking-related lung cancer share: 80-90% - Taubes says most lung cancer cases occur among smokers.

Pivotal Quotes: "Yeah, virtually everything. The short answer." — Gary Taubes: His blunt response to the claim that ‘eat less, exercise more’ explains weight loss. "We actually solved the obesity epidemic in the 1960s." — Gary Taubes: Taubes argues the correct hormonal understanding existed decades ago but was discarded. "The appetite comes with the meal." — Russ Roberts: Roberts summarizes the idea that eating can stimulate hunger rather than merely relieve it.

Implications: If Taubes is right, weight-loss advice should prioritize carbohydrate quality/quantity and individual metabolic response, not just calorie reduction or exercise. Nutrition research and public health guidance may need major revision.