Food stamps piss people off. They are a convenient way to confirm existing prejudices, typically a none-too-subtle dog whistle for racial stereotypes about the poor – who they are, what they look like, and the content of their character. Like many of the things that trigger people emotionally, food stamps are resonant, symbolic anchors for anger that allow people to condense complicated, multi-layered social experiences like poverty into warrants and judgments on the character and moral worth of individuals. … the reduction of all experience to a social vacuum – we are simply
And it is true that the food stamp program has problems. Reports of spending on soda and other junk food. resistance to change from USDA, intense lobbying from soft drink companies. The politics of food in the US are always messy. Classically true, also, that food options for those with little money are limited and tend to be less healthy. Subsides are not huge, but for many all that keep people off the street.
Work issues. it is true that low-wage companies (ironically, often food companies and restaurants that serve the people who are least wealthy) benefit from food-stamp subsidies. Many working people cannot make it without food stamps.
Fraud concerns. all red herrings. degree
What is the reality?
Widely known that the Obama Administration loosened food stamp eligibility standards after the financial meltdown of 2008. Between 2005 and 2014, the number of households in the US increased by 5.47 percent, from 112 million to 118 million. In that same period, the number of households receiving food stamps increased by 72.17 percent, from 9.3 million to 15.9 million. Food stamp eligibility requirements differ dramatically across states, and so the opportunity for working with this data is to take the increase as a given and instead to unpack the relative impact of this increase on different population groups, specifically breaking down by race, age, and poverty status.
What this means, specifically, is a two-step methodology, in which the change in household numbers, for each demographic unit in each state, as a percentage of the change
We also can isolate trends associated with strongly partisan Red States and Blue States, defined as states with at least a 15 point gap separating Donald Trump and Hillary Clinton in the recent presidential election. For example, in Red States, food stamp recipients increased by 35 percent between 2005 and 2014, while in Red States households receiving food stamps increased by 98 percent. For our purposes, it is not important what economic circumstances or policy decisions led to these divergences. Mapping the flow of food stamp benefits within states and within these political geographies is our only concern. We will consider age, poverty status, and race in turn. The results may surprise you!
- Key Variables – The analytic pivot is the relationship between food stamp recipients for a given population and the absolute size of that population (e.g., the percentage of households with children under 18 that receive food stamps in relation to the total number of households with children under 18). However, we want also want to capture the dynamic relationship between food stamp recipient numbers for different population groups as they change over time (e.g., how do changes in the percentage of households with children under 18 that receive food stamps compare to changes in the percentage of households with members over the age of 60 compare). The best way to capture these two relationships is by separately establishing the proportional weight of total populations and the proportional weight of food stamp recipients within those populations and then comparing those proportions as a ratio to establish a relative weight. Here’s an example. In Washington State, there were a total of 2,450,474 households in 2005. Of that total, 813,557 were households including children under the age of 18. So 33.2 percent of Washington State households included children under the age of 18 in 2005. In this same year, 207,478 households in Washington State received food stamps. Among these households receiving food stamps were 107,889 households receiving food stamps that included children under the age of 18. So 52.0 percent of the households receiving food stamps included children under the age of 18. This means that a disproportionately high number of households with children under the age of 18 received food stamps relative to the entire population of households in the state. To be precise, the ratio of 52.0 percent for food stamp households to 33.2 percent for total households that included children under the age of 18 yields a proportionality weight of 56.63 percent. In other words, if an average yield of food stamp recipients from a set of populations was 100, the average yield for households with children under the age of 18 would be 156.
- Young and Old – In keeping with the spirit of most social welfare programs, food stamps primarily target the nutritional and food security needs of children. However significant numbers of households with at least one person older than 60 also receive these benefits, and between 2005 and 2014, recipient trends skewed toward households with older members and away from households with children under the age of 18. Households with older populations increased by nearly 30 percent in the United States in this time frame. while households with children under 18 decreased by 4 percent. At the same time, food stamp benefits for older households more than doubled from
Compared to households with children under 18 years of age, note the disproportionately large percentage increase of households with members over 60 years old between 2005 and 2014, and the even more dramatic surge in food stamp recipients. Older households are proportionally less likely to be on food stamps and households with children under 18 are proportionately more likely to be on food stamps. But the gap narrowed between 2005 and 2014.
Comparing food stamp growth relative to states going hard blue or red in the 2016 election (at least a 15 percent separation between Trump and Clinton), the most important development is the growth in Blue State food stamp recipients proportionate to population growth, particularly for the older 60 household cohort.
use bullet points to make lots of small assertions based on data
hold increase stable – focus on changing mix over time, by group – link to migration patterns, changes in identity labels, less the result of politics than a way to frame and understand politics – look at impact of group population growth rates
methodology – large percentages of households/recipients versus small percentages as 2005 baseline
Families with 1 worker in past 12 months – almost as dependent on food stamps as families with no worker in the past 12 months
Families with 2 or more workers in past 12 months – doing just fine, although also a bit of relative decline
do relative increase in minority benefits occur more in Trump states or relative decline in white benefits?
whites in DC – gap with blacks – significance of large gaps between groups – higher population percentages should bring group closer to 0 baseline (e.g., Vermont) – but not always true (DC)
absolute percentages – account for variance (e.g., whites in West Virginia v. Wisconsin – both of which went for Trump)
super red states significantly more white, and more stable in their percentages – even though higher percentage on food stamps
blue states – outsized impact of California
there are many ways to focus – will focus on race, children
impact of economic growth patterns, changes to eligibility laws unclear – need to hold constant across states (?)
- WHITE – not much household growth – balancing change when household growth static versus change when household growth dynamic
- ASIAN – crushing it, but not uniformly
- BLACK – California / Delaware
- MIXED RACE
WORK – families with working age populations (difference from total household population ratios) – a significant help for families without workers – assess change over time
ideas about basic income, as way to assess impact – assessing changes in income in distributional terms
also, macroeconomic recovery and food stamps
relative income – Delaware, DC, Idaho (but consider also in relation to percentage of population that are recipients) – also indicates that more difficult stay afloat with higher income? or that a function of change in eligibility (loosening?) – but again, consider relative differences (which are minor?)