Can taking a quiz influence people’s opinions about whether taxes are fair? To answer that question, Tax Policy Center scholar Vanessa Williamson ran an experiment with the news website Vox. What she learned is that tax facts matter. Drawing people’s attention to different information about taxes significantly shifted their attitudes about who should pay how much and even whether their own taxes are fair.
Here are three key takeaways from the experiment:
Facts about taxes on lower-income people matter most. Getting people to think about the taxes that lower-income people pay affected their attitudes more than getting them to think about taxes that higher-income people pay.
Fairness works like a see-saw. If people think lower-income people pay too much tax, they tend to think that higher-income people pay too little.
Ignorance might be bliss. Answering any questions about taxes made people think their own taxes were less fair.
Who Took the Quiz?
7,710 participants from every state, the District of Columbia, and Puerto Rico
Most quiz takers lived in urban areas
76 percent were men
Based on Vox readership, quiz takers were likely younger, more politically active, and more liberal than average
The Experiment
In February 2017, over 7,000 Vox readers took a short online quiz about taxes. Within that quiz was an experiment: at random, people answered one of four sets of factual questions about taxes. Each set of factual questions suggested a different perspective:
People with higher incomes pay relatively low taxes
People with higher incomes pay relatively high taxes
People with lower incomes pay relatively low taxes
People with lower incomes pay relatively high taxes
Respondents then answered five questions concerning their opinions about tax fairness:
Do higher-income people pay too little, too much, or the right amount?
Do lower-income people pay too little, too much, or the right amount?
What’s the highest tax rate anyone should pay?
Should higher-income people pay more taxes to redistribute incomes?
Are your own taxes fair?
To help assess changes caused by the different factual questions, we created a control group, who were asked the opinion questions before getting a set of factual questions.
What Did We Learn?
Thinking about tax facts can change people’s views about tax fairness.
Above all, the Vox quiz experiment showed that focusing respondents’ attention on particular aspects of the tax code had sizeable effects on their attitudes about the distribution of tax liability and on their perceptions of their own taxes’ fairness.
Vanessa Williamson, “What Makes Taxes Seem Fair?”
Asking questions about taxes paid by lower-income people shifts people’s views about whether those with lower incomes are taxed fairly.
About 56 percent of people in our control group—those who were asked their opinions before taking the quiz—said that lower-income people pay too much tax. That percentage goes up by about one-fifth if people are first asked questions suggesting that lower-income people pay relatively high taxes. Conversely, the percentage goes down by about one-sixth if people are first asked questions suggesting that lower-income people pay relatively low taxes.
Fairness works like a see-saw. Asking questions suggesting that higher-income people pay low taxes or that lower-income people pay high taxes both make people feel that those with higher incomes should pay more.
In our control group, 82 percent said that higher-income people pay too little in taxes. That percentage increases modestly if people are first asked questions suggesting that higher-income people pay relatively less in taxes. The percentage increases about the same amount if people are first asked questions suggesting that lower-income people pay relatively more in taxes.
Ignorance might be bliss. Asking questions about taxes paid by others, whether focusing on people with high or low incomes, leads people to think their own taxes are not as fair.
About 62 percent of people in our control group said their own taxes were “somewhat fair” or “very fair.” Seeing any questions about taxes lowered that percentage: It fell by nearly one-fifth if people were first asked questions suggesting that lower-income people pay relatively high taxes. It fell by smaller amounts if the questions suggested that high-income people pay relatively low or relatively high taxes or that low-income people pay relatively low taxes.
