Conflicts over Federal government Spending have been a defining feature of 21st-century American politics. It is not surprising that a general uptick in opposition to spending should follow a global economic crisis like the Great Recession. But the dynamics of the conflict are curious — in particular, fierce critiques of spending have come from areas of the U.S. that rely more heavily on federal money.
In a paper recently published in American Politics Research, we document and analyze this phenomenon, which we call the federal spending paradox. We began by calculating each state’s ratio of spending to taxes, dividing the average amount of federal money received by each state per year from 2001 to 2010 by the average taxes paid by each state per year over that decade. We used data on spending from the Census Bureau’s Consolidated Federal Funds Report and data on taxes from the Internal Revenue Service. States with spending-tax ratios below 1.0 received less money from the federal government than they paid in federal taxes (e.g., Delaware got $0.42 per tax dollar paid). States with ratios above 1.0 received more money from the federal government than they paid in federal taxes (e.g., New Mexico got $2.91 per tax dollar paid).
Looking at the relationship between this ratio and the percentage of each state’s population that opposed spending (taken from the 2010 Cooperative Congressional Election Study), we found a positive and significant correlation: A 10-percentage-point increase in opposition to Federal Spending is associated with an additional $0.37 in federal outlays per tax dollar paid, or an extra $17.1 billion in federal money for a state with an average tax burden.
Why is opposition to federal spending higher in states that receive more federal money? Many pundits have argued that poor people rail against the very spending on which they rely, and more specifically that poor Republicans in so-called “taker” states accept government benefits with one hand while using the other to hold rally signs disparaging Democrats in so-called “giver” states for excessive spending. This argument may sound intuitive, since support for Republicans and opposition to spending are both higher in taker states than in giver states — but it is not supported by evidence.
The state-level spending paradox we observe is not a simple product of economic irrationality. When we looked at the national population to examine opinion on federal spending at the individual level, we found sensible relationships between people’s economic circumstances and their support for government spending. Characteristics that make people more likely to benefit from government spending (e.g., having lower income, being unemployed, having a disability, and receiving government health insurance) make them more likely to support government spending, even accounting for partisanship, ideology, and other characteristics.
This deepens the question — if the relationship between economic status and opinion on spending makes sense at the individual level, how could it be paradoxical at the state level? We tackled this puzzle by first analyzing noneconomic influences on individuals’ fiscal attitudes. Using data from the 2010 Cooperative Congressional Election Study (CCES), we looked at the impact of three factors in particular: party identification, ideology, and racial resentment. Also known as symbolic racism or modern racism, racial resentment is a post-Civil-Rights-era view rooted in the denial of continuing discrimination against African Americans, doubts about their work ethic, and resistance to government efforts to reduce racial inequalities. The CCES asks questions that measure these views.
As one might have guessed from the racial undertones often present in public discussions on fiscal politics, greater racial resentment was associated with lower support for spending. This remained true even when we accounted for other demographic and political characteristics, such as gender, race, age, education, income, party identification, ideology, and so on. In fact, racial resentment was far more powerful in predicting opposition to federal spending than economic self-interest was — for example, it was four times stronger than income. Its influence exceeded even that of party identification, which is notable in our era of hyperpartisanship. So economic characteristics do matter in the way we would expect; it’s just that other factors matter more.
Then we examined how individual opinion on spending aggregated to form state-level opinion. The relationship between factors like income and opinion on spending can look different at the individual and state levels for two reasons. First, the same characteristic may influence people’s opinions differently in different places. People weigh various interests when formulating opinions on issues — e.g., considerations relating to their gender might push them in one direction, while those relating to their income might push them in another. Political culture, which varies considerably across states, can shape how people prioritize and balance these different concerns in developing their positions on issues. The key question is whether the power of certain factors known to predict opposition to spending (like higher income and racial resentment) varies systematically across giver and taker states.
Second, people are not randomly distributed across states. Even if we assume the impact of a particular characteristic is the same in every state, there may be more people with that characteristic in some states than in others. For instance, if there are more people with racial resentment in state A than in state B, all else equal, average opposition to spending will be higher in state A. So the other critical question is whether there are more people with characteristics associated with opposition to spending in taker states than in giver states.
We found that the aggregation of individual opinion to state-level opinion contributes to the federal spending paradox in two ways. Income has a stronger impact on opinion about spending in taker states than in giver states. In fact, in many giver states (e.g., Delaware, New Jersey, New York, and Massachusetts) the relationship between income and opinion on spending is weak to negligible. This means, for example, that wealthy people in Delaware do not oppose federal spending much more than lower-income people in Delaware, while wealthy people in New Mexico are significantly more opposed to federal spending than lower-income people there. In the case of racial resentment, more people harbor racial resentment in taker states than in giver states, raising average opposition to spending in taker states and contributing to the spending paradox.
In the end, there is nothing strange or pernicious about giver and taker states. It is common in liberal democracies to redistribute wealth across regions. But the federal spending paradox creates difficult mandates for many members of Congress representing poor states. Does good representation mean listening to constituents’ opinions, even if doing so would hurt the state economically? This question is further complicated by the powerful role of racial resentment in shaping people’s opinions on spending. With budget battles, proposed tax reforms, and cuts to major government programs looming on the horizon, it has become even more important to think about these issues.