The proposed insurance hypothesis is intriguing but seems to fall short of addressing a number of issues that contribute to obesity levels not only in adult women, but also in all persons from affluent and less affluent societies. For a start, the proposed hypothesis overlooks the role of psychological determinants of behaviours that lead to obesity such as taste, pleasure, mood, and habitual responses. There is compelling evidence that these factors impact eating behaviour and thus obesity (Gibson Reference Gibson2006). The insurance hypothesis suggests that food consumption is driven by survival, yet much research suggests that in humans, food is consumed because of taste and pleasure rather than survival (e.g., see Jansen et al. Reference Jansen, Theunissen, Slechten, Nederkoorn, Boon, Mulkens and Roefs2003). An additional factor that affects food consumption regardless of satiety consists of the habitual affordances of the environment. For example, Neal et al. (Reference Neal, Wood, Wu and Kurlander2011) found that people consumed significantly more stale popcorn in an environment that cued the behaviour (a cinema), compared to an environment where there was not the same cue to action (a classroom). This line of evidence would appear to be counter to the insurance hypothesis.

A second significant contributor to obesity not captured by the insurance hypothesis is the negative energy balance experienced by most people living in the industrialised world. For example, energy intake (kcal/day), as recorded in the National Health and Nutrition Survey by the Centers for Disease Control and Prevention from 1971 to 2000, shows no noticeable change across time in United States women ages 20–74 years. In contrast, levels of physical activity (occupational, household, and transportation) have declined substantially (Archer et al. Reference Archer, Shook, Thomas, Church, Katzmarzyk, Hebert, McIver, Hand, Lavie and Blair2013). In the United Kingdom, the Institute of Fiscal Studies (Griffith et al. Reference Griffith, Lluberas and Luhrmann2016) found that total calories purchased from 1980 to 2013 have actually decreased. However, levels of obesity in the United Kingdom have increased because of substantial decreases in strenuous physical activity at work and in daily life. Such decreases in strenuous physical activity have been accompanied by substantial increases in sedentary behaviours; data indicate that adults in high-income countries spend most of their time sitting at work or watching TV (e.g., Matthews et al. Reference Matthews, Chen, Freedson, Buchowski, Beech, Pate and Troiano2008). Hence, reductions in energy expenditure contribute to obesity in addition to food availability and perceptions of food insecurity.

A third factor that contributes to the obesity problem, but not captured by the insurance hypothesis, is the built environment. Availability and accessibility of infrastructure for walking and bicycling, perceived safety, and aesthetic attributes have been found to predict levels of physical activity and obesity (Sallis et al. Reference Sallis, Floyd, Rodriguez and Saelens2012). Thus, the built environment, in conjunction with the food environment, contributes to obesity. A food environment in which there is easy geographic access to fast-food outlets and convenience stores encourages individuals to consume foods that are high in energy and saturated fats. In fact, a recent literature review showed greater availability of fast-food restaurants in low-income neighbourhoods (Fraser et al. Reference Fraser, Edwards, Cade and Clarke2010). Pervasive food marketing exacerbates the effects of the obesogenic environment. In developing countries, where as much as 60% of household income is spent on food (Caballero Reference Caballero2007), marketing campaigns and price incentives for high-calorie products have a substantial impact on food-purchasing patterns.

Finally, energy-dense, nutrient-poor food is accessed easily worldwide and not just in high-income countries. Evidence shows that in low-income countries, such as India, Vietnam, Bolivia, and Nigeria, energy-dense nutrient-poor fast food is becoming increasingly popular (Rockefeller Foundation 2013). As an example, in India between 2007 and 2012, consumption of soft drinks increased by 70% and consumption of unhealthy foods by 110%. Counter to the proposal that the association between food insecurity and high body weight is restricted to adult women from high-income countries, women in the developing world often eat the least nutritious food for several reasons: lack of time, poverty, or because they are the last in the family to eat (Rockefeller Foundation 2013). Thus, data from the International Association for the Study of Obesity (Rokholm et al. Reference Rokholm, Baker and Sørensen2010) show that high body mass index (BMI) rates greater than 30 in adult women can be found in both affluent (e.g., United States, England) and less affluent countries (e.g., Samoa, Egypt, Mexico). Conversely, current data from the United Kingdom suggest that increases in obesity have slowed (Sperrin et al. Reference Sperrin, Marshall, Higgins, Buchan and Renehan2014) at a time when food insecurity is increasing (Loopstra et al. Reference Loopstra, Reeves, Taylor-Robinson, Barr, McKee and Stuckler2015), an observation that would appear to directly contradict the insurance hypothesis.

In summary, the insurance hypothesis seems insufficient to account for increases in obesity, as it ignores a number of important contributing factors, such as psychological factors affecting eating behaviour, the reduction of physical activity leading to a negative energy balance, and changes to the built environment in which we live. Moreover, this hypothesis seems incompatible with current data on obesity, particularly in low-income countries.