Mike Konczal has written a really intriguing post on consumption inequality, with a focus on “upscale” versus “downscale” food consumption. And he cites Christian Broda and John Romalis, two economists who’ve done a great deal of work on globalization and the gains from variety. In Prices, Poverty, and Inequality, Broda and David Weinstein argued that we can’t understand the U.S. consumption landscape without understanding that the effective inflation rates experienced by more and less affluent households are very different. A big part of the story more affluent households consume more labor-intensive goods, in part because higher wages create an incentive to outsource more household labor. Mike doesn’t address the Broda and Weinstein argument re: services, which is a shame.
Rather, he focuses on food to suggest that the dietary preferences of the rich and the poor, shaped and constrained by food prices, point to wider concerns with the Broda-Romalis argument. As Mike notes, Broda and Romalis spend a lot of time talking about food. And he makes an excellent point:
The poor have more purchasing power because, in part, they are buying food that isn’t very healthful. And the important thing about this different inflation rate quantification for income inequality is that nobody gets diabetes. The long-term health costs of “choosing” a different inflation rate for your food isn’t estimated, nor are they included to see if it all balances out economically.
Which is to say that when you hear arguments that income inequality isn’t so bad because consumption inequality is less than you’d think, it’s important to be skeptical about what in fact is being consumed. Sometimes it is exactly what you think it is: less quality and worse long-term health outcomes. And the long-term consequences for the health and well-being of the working poor are exactly the type of information econometric stats obscure.
But I’m not sure I get the upshot here. I think I’m supposed to take away that “consumption inequality is less than you’d think” because the poor are eating deadly, unhealthy food, which generates misery and high health expenditures. But we’re talking about the impact of differential inflation rates on individual consumption. If we want to factor in the costs of managing diabetes, we could certainly factor in the pain and suffering associated with the associated pain and the risk of early death. We could also factor in the fact that care for the poor is heavily subsidized through progressive taxes, and that the high health expenditures are borne in no small part by the affluent. Should we thus conclude that the reckless eating habits of the poor are creating a terrible burden on upper-middle-class taxpayers? That doesn’t sound right to me, though of course this sentiment could account for the food paternalism that’s become increasingly popular among affluent Americans. Lest we forget, it is also possible that unhealthy habits that shorten lifespans actually reduce health expenditures overall. My guess is that Broda and Romalis steered clear of this thicket because it would lead them into an endless morass of assumptions and counter-assumptions.
Much depends on the nature of our heterogeneous preferences. An unhealthy diet could shorten one’s lifespan while also giving one a more pleasurable life. For someone else, the shame and stigma associated with obesity could be crippling, and the narcotic effect of eating fatty foods could serve as a kind of self-medication. That really does sound like an awful trap.
I wonder if dividing people into income quintiles is the best way of thinking about the social patterns surrounding food consumption. For example, there are at least some shoppers at Whole Foods who aren’t very affluent, but who place a very high premium on purchasing what they imagine to be high-quality food. To that end, they will economize in other domains, e.g., clothing, entertainment, etc. The willingness to delay gratification in this way could reflect habits and expectations that facilitate upward mobility.
It is possible that subsidizing fresh fruits and vegetables would lead to a sharp spike in consumption of fresh fruits and vegetables. It is also possible that many people — I think I’m one of them — prefer low-quality animal protein and sugary drinks to the healthier alternatives. One could characterize this as hyperbolic discounting of the worst kind, an indifference to the long-term costs of eating less-than-healthful foods. I wonder how susceptible these preferences are to subsidies and other interventions.
One way to express what I take to be Mike’s lament is that rising raw wage dispersion — quite possibly an artifact of a broken health system — means that poor people are forced to buy cheap, delicious food that is killing them. Another way to express it is that poor people should eat better food and spend less on technologies that make environments more obesogenic, like televisions and air conditioners.
Consider the following findings from “Putative contributors to the secular increase in obesity: exploring the roads less traveled” in The International Journal of Obesity:
Regarding vending machines, a thorough evidence-based review (MS Faith et al., unpublished, 2005) found no published randomized trials, quasi-experiments or observational epidemiologic studies evaluating their effects on obesity. Regarding fast-food availability, although some studies showed associations with obesity, Burdette and Whitaker found no association between being overweight and proximity to fast-food restaurants in over 7000 children. Regarding HFCS, the leading source (in the United States) is sweetened beverages and three out of four studies conducted in children have found no association between soft drink consumption and BMI when controlling for total energy intake, raising the issue that there is no independent effect of HFCS calories on body weight, other than its pleasant taste possibly leading to the potential increase in total caloric intake as would any food.
The authors suggest a number of other drivers, including sleep debt, reduction in the variability of ambient temperature, and increasing gravida age. Given that the less affluent tend to work shorter hours than the more affluent, per Aguiar and Hurst, sleep debt should be an easy problem to tackle, except sleep competes with a number of other increasingly attractive entertainment options. Reducing the use of HVAC systems could prove very beneficial, and it’s not obvious that reducing raw wage dispersion would help matters. And I have no idea as to how we might tackle increasing gravida age.
Overall, I think that we overestimate the importance of food deserts and wage dispersion in causing public health problems and we underestimate the importance of encouraging the consumption of natural butterfat and animal fat when it comes to solving them. But that’s just me.