Poor Choices or Poor Analysis?

fA Critical Review of Food-Based Poverty Traps

There’s an interesting tension between libertarianism and paternalism when we talk about the lives of the poor. One the one hand, certain analyses of poverty take a rather patronizing view of the poor. These approaches often feature some exogenously-defined “rationality” criteria which are supposed to advance the lives of the poor, who are then seen as unable to meet these criteria due to various biases and impairments of judgment. For example, if only the poor were to consume more calories, they could work more and earn more money. Instead, they spend their income on cigarettes. In this narrative, the poor do not know what’s good for them, so the wise researchers from (mostly) Western academia must educate them in how to live correctly. These approaches, their good intentions notwithstanding, may stigmatize poverty and miss the relevant drivers of choice by poor households. On the other end of the spectrum, there are the more libertarian approaches that seek to give theoretical agency to the poor, albeit at the expense of ignoring context and circumstance. These analyses run the risk of overly romanticizing the poor as “resilient and creative entrepreneurs,” forgetting that many do not actively choose the life of subsistence entrepreneurship, for example. In other words, by reframing all of choices made by poor people as “rational” choices, these approaches leave no room for envisioning qualitatively-better choice sets. I believe that we need to be careful in how we talk about and conceptualize the lives of the poor. On the one hand, we really need to paint the big picture of why poor people choose what they choose — beyond our own biases of “good” and “bad” choices — to understand what needs these choices may serve. This would give more agency to the choices that poor people make, and constitute the healthy libertarianism necessary for viewing poor people as more than just biased victims of poverty. On the other hand, we cannot give up on normative judgments if we want to help the poor improve their living conditions. The poor may have good reasons for why they make the kinds of choices they do, but good analysis and sound policies may serve those reasons better.  

An example might illustrate what I mean: Most literature has treated cigarette consumption by the poor as an addiction, an unwanted habit that consumes an undue portion of income. If this income was instead spent on acquiring more calories, the poor could increase their productivity and improve their standard of living, or so the story goes. However, these approaches ignore the possibility that cigarettes may serve a very direct and relevant need – the reduction of stress levels caused by living on the margin of survival. Thus, a more libertarian approach would recognize that in addition to feeding themselves, the poor are correctly concerned with stress-reduction, which is necessary for maintaining or indeed increasing their productivity. By narrowly focusing on calorie intake, and treating cigarette consumption in a rather patronizing manner, these approaches generate bad analysis. However, the other extreme would be acknowledging that poor workers use cigarettes for stress-reduction, and empowering this choice in the name of “freedom of choice.” The problem here is that cigarettes kill. Therefore, a true aid to the poor would be the provision of alternative, healthier methods of stress-reduction that can compete with cigarettes in terms of effectiveness and affordability. Notice that such a policy can only come about by a mixture of libertarianism and paternalism: first, we should take the choices of the poor at face value to understand what needs they serve; and once we know this we must seek to provide better, healthier ways of meeting these same needs. Instead, researchers have too often focused on what the poor should need, and then wondered why they don’t behave accordingly. 

To make my arguments, I will primarily rely on the essay by Banerjee & Duflo (2006) called “the economic lives of the poor.” Through this and other closely associated texts I will try to establish how development economics tends to approach the issue of nutritional spending by the poor. Banerjee & Duflo analyse the relationship between income, nutrition and productivity as follows: people spend their energy on work, which earns them income; they use their income to purchase food, which gives them energy to perform work tomorrow. Thus, food (or energy received from food) mediates the relationship between income today and income tomorrow. While this model implies that the poor should consume as many calories as possible, this is not actually the case. More specifically, the poor very frequently spend their money on “expensive calories” such as sugar, or non-food items such as cigarettes and tobacco.

I will challenge Banerjee & Duflo on the following account: I will argue that the focus on caloric intake is too narrow, and the link between calories consumed and productivity is exaggerated at the expense of other highly-relevant causal factors. Specifically, expanding the model by incorporating “stress” explains observed behavior better than the original model. Using some evidence from neuroscience, I will show that expenditures on sugar, alcohol and cigarettes are actually quite predictable if we understand that productivity is related to stress. Stress affects cortisol levels, which affects blood-sugar levels and causes craving for sugar. Similarly, both alcohol and tobacco are shown to regulate cortisol levels (albeit at the expense of other health factors). Thus, the poor are actually making choices that help them maintain their productivity, but through channels that are ignored by the authors. 

In the final analysis, I will complement this “healthy libertarianism” with a “healthy paternalism” by arguing that just because the poor use sugar, tobacco and alcohol for stress-reduction does not mean that it’s the best idea. These substances have long-term detrimental effects that are not fully internalized by individuals. Thus, it is in the long-run interest of the poor to have access to stress-reduction tools that are just as affordable and effective as tobacco or alcohol, but are less damaging. It is completely beyond the scope of my knowledge to promote any actual remedies, so my contribution is merely showing that we can talk about the poor in a less paternalistic way while also improving our analysis.

One thought on “Poor Choices or Poor Analysis?

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s