Nettle et al. propose the insurance hypothesis to explain the empirical relation between food insecurity and body overweight. The results of their meta-analysis suggest that this relation is significant only among women in high-income countries. The focus of the model on survival motivations falls short of explaining these empirical findings. More importantly, the insurance hypothesis might not explain the data better than alternative psychological mechanisms that are relevant to food consumption and health. To demonstrate this point, we suggest two mechanisms as potential alternative explanations: (1) temporal discounting and (2) cognitive implications of resource scarcity.

Temporal discounting (Frederick et al. Reference Frederick, Loewenstein and O'Donoghue2002) explains seemingly maladaptive behaviors like gaining excess weight by an individual's tendency to be driven by immediate rather than by future costs and benefits (Hall & Fong Reference Hall and Fong2007). Such a short-term view suggests, for example, that families living on welfare allowance might consume most of their budget right after receiving their food stamps, forcing them to shift later to consuming low-price food that is calorie laden and unhealthy (Drewnowski & Darmon Reference Drewnowski and Darmon2005). A natural reason for this myopic view and future discounting is that the future is more uncertain than the present. In the food stamps example, unforeseen events might reduce their net benefits (e.g., decrease the stamps’ purchasing power), so it might make sense to use them right away. Recent experimental studies have yielded empirical support for this effect of uncertainty on temporal discounting (Epper et al. Reference Epper, Fehr-Duda and Bruhin2011; Milkman Reference Milkman2012). In the context of Nettle et al.’s analysis, these observations suggest that food insecurity might affect body overweight by facilitating temporal discounting, rather than by activating an “insurance” mechanism against shortage in food supply. The temporal discounting hypothesis does not necessarily predict a gender effect in the data that showed the effect of food insecurity on overweight mostly among women (much like the insurance hypothesis, which requires additional assumptions for explaining this pattern). Yet perhaps a first clue can be suggested by recent findings showing that although obese women exhibit greater temporal discounting than healthy-weight women, this pattern is not observed among men (Weller et al. Reference Weller, Cook III, Avsar and Cox2008).

In any event, the current empirical data cannot distinguish between these two very different, even somewhat opposing, mechanisms. Temporal discounting suggests that food insecurity facilitates a short-term view and future discount, whereas the insurance hypothesis suggests that it actually activates a long-term view (insurance against future shortage). More empirical data are needed to test which of the two mechanisms operate in the context of food insecurity.

We now turn to another alternative explanation for the food insecurity – obesity relationship: the psychology of scarcity. Recent evidence suggests that whenever people feel that resources are low relative to their needs, a scarcity mind-set emerges, and it changes the process of making decisions (Shah et al. Reference Shah, Mullainathan and Shafir2012). Specifically, under conditions of scarcity, pressing needs capture attention (Aarts et al. Reference Aarts, Dijksterhuis and Vries2001; Shah et al. Reference Shah, Mullainathan and Shafir2012). For example, income deficit elicits greater focus on expenses, and hungry or thirsty individuals focus on food- and drink-related cues (Aarts et al. Reference Aarts, Dijksterhuis and Vries2001; Radel & Clément-Guillotin Reference Radel and Clément-Guillotin2012). The attentional focus on satisfying scarce resources, mainly those who are needed to assure functionality, might result in boomerang effects. For example, low-income individuals tend to take loans with high interest that help in meeting today's needs, focusing on the loan's benefit rather than on its costs. Yet paradoxically, taking these loans makes future expenses harder to resolve. Food insecurity could be described as a natural form of scarcity, which might lead people to focus on short-term benefits of eating (energy, taste), rather than on its costs (excessive weight), and consequently yield excessive food consumption. In addition, resource scarcity was found to yield poorer self-control (Mani et al. Reference Mani, Mullainathan, Shafir and Zhao2013), which also facilitates short-term behavior and excessive consumption.

Another form in which the feeling of scarcity might affect food consumption relates to time scarcity, which might act not only as a cognitive barrier, but also as an “objective” barrier to healthy food consumption and exercise (as both typically require some investment of time; Jabs & Devine Reference Jabs and Devine2006). Given the lifestyle of modern society, time scarcity seems like a natural account for explaining why the food insecurity–obesity relationship was found mostly in high-income countries. In high-income countries, time scarcity coupled with the high availability of ready-to-eat food is associated with increased likelihood of obesity (Jabs & Devine Reference Jabs and Devine2006). Furthermore, one might also speculate that time scarcity might also yield the gender effect observed in the meta-analysis data, as women in modern societies tend to experience higher time scarcity than men (Bittman & Wajcman Reference Bittman and Wajcman2000; Mattingly & Blanchi Reference Mattingly and Blanchi2003). Once again, the main point we would like to emphasize is that the current data are not rich enough to differentiate between the insurance hypothesis and the scarcity mind-set hypothesis, which suggest very different mechanisms.

Nettle et al.’s data are intriguing but are inadequate for differentiating between the alternative mechanisms we propose here and the insurance hypothesis. One could possibly think of other processes of relevance, such as the projection bias referring to the tendency to overpredict the degree to which future needs and tastes will resemble current ones (Loewenstein et al. Reference Loewenstein, O'Donoghue and Rabin2003). It is also plausible that several processes operate in the current context and interact with environmental variables (e.g., culture, economy) and/or with individual differences (e.g., gender, age). We share the authors’ enthusiasm regarding the intriguing relationship between food insecurity and overweight. Moreover, we see much potential in their evolutionary approach. Nevertheless, the question about the exact mechanisms that give rise to this relationship remains open to different interpretations. Additional data collection and experimental analyses are needed for better understanding of the actual forces that drive this interesting connection between food insecurity and overweight.