By Michael Cummings
Yale economist Ebonya Washington draws her research ideas from the real world. A newspaper article, an observation while riding the bus, or an interaction at a store can raise questions worthy of analysis, she says.
Not long after she joined Yale’s faculty in 2004, Washington was standing in line at a supermarket in New Haven when a potential idea for a paper presented itself.
“It was a very long line,” she said. “The woman in front of me turned to me and said, ‘Don’t you just hate the way they raise the prices the day that food stamps come out?’”
Washington decided to test the woman’s assertion. A paper she published with economist Justine Hastings, a Yale colleague at the time, analyzed scanner data from a national supermarket chain. They found a small increase in prices after monthly benefits were issued, corresponding to a larger increase in demand. As a potential remedy, they suggested staggering the issuance of benefits to prevent all recipients from receiving them on the same day.
Washington, the Samuel C. Park Jr. Professor of Economics, specializes in public finance and political economy — the study of how political factors influence economic outcomes.
“I’m largely interested in how people use politics to get their economic needs met, particularly people of color and low-income folks,” said Washington. “A nice thing about being an academic is that you have the ability to pursue questions that interest you. I think we can use economics to answer really interesting questions about our society. A lot of people don’t realize that.”
Understanding political motivations
At times, those questions require testing conventional wisdom. Last year, Washington published a paper with Princeton University economist Ilyana Kuziemko examining the role that racial attitudes played in driving southern whites from the Democratic Party to the Republican Party during the 1960s.
“Many people believe that race was a big factor in why the Democrats lost the South, but there wasn’t much quantitative data available to support that belief,” Washington said. “In the scholarly literature, there was a quantitative argument that economic development caused southern whites to abandon the Democratic Party as we tend to associate higher-income people with Republican policy proposals.”
Those quantitative studies did not find a role for race in the party realignment because of the lack of good data measuring people’s attitudes about race before and after the Democrats proposed major civil rights legislation in the 1960s, she said.
By analyzing little-used Gallup survey data that asked people beginning in 1958 whether they would vote for a well-qualified black candidate for president, Washington and Kuziemko determined that racial issues were the key factor in the South’s political reorganization. They corroborate their finding with an analysis of newspaper coverage of President John F. Kennedy on civil rights issues, which allowed them to show that Kennedy’s approval rating among Southern whites dipped in each time period in which there was an increase of articles linking his administration to civil rights in the South. The support cratered in June 1963 when he announced his support for sweeping civil rights legislation. Although conventional wisdom points to President Lyndon B. Johnson losing the south with the signing of civil rights legislation, their findings point to Kennedy as an initial cause.
“It’s important to understand people’s political motivations,” Washington said of the study. “It helps us to understand what people consider important to predict what issues are likely to influence future elections.”
Washington has also studied the impact of the Voting Rights Act of 1965, a landmark legislative achievement of the civil rights movement. In 2014, she published a study with Dartmouth College economist Elizabeth Cascio analyzing the effect of expanded enfranchisement on the amount of state funding distributed to black communities in states where literacy tests had previously denied black citizens their voting rights.
They found that counties with higher shares of the black population saw a relative increase in state funding transfers. For each 10% increase in a county’s 1960 black population share, the removal of literacy tests caused nearly a 6% increase in per capita state transfers during the 15 years after the Voting Rights Act was enacted, amounting to a 16.4% increase in per capita transfers over the period, according to the study. The majority of the state funding supported public education, Washington said.
“Again, people could probably assume that expanded voting rights have benefited black people, but quantifying those benefits is important,” Washington said. “Doing so helps to inform policy moving forward, especially at a time when the Voting Rights Act is in danger of being gutted even further.”
‘The Mommy Effect’
Two studies Washington is currently working on show the breadth of her research interests. One explores why the gender gap in employment has plateaued in wealthy countries despite the fact that women acquire more education than men. The other investigates whether people’s attitudes toward government cause them to evade federal income taxes.
The first paper, “The Mommy Effect: Do Women Anticipate the Employment Effects of Motherhood?”, asserts that women miscalculate the difficulty of balancing work with raising a family when they make decisions about their educations. Using data from the United States and the United Kingdom, Washington and her co-authors document a large decrease in female employment at the time of the birth of the first child. They argue that this decrease is unanticipated. One of the key pieces of evidence comes from survey questions on attitudes toward combining motherhood and employment. For example, the British Household Panel Survey asks for agreement with the statement “All in all, family life suffers when a woman has a full-time job.” At the time of the birth of the first child women become more conservative — less likely to believe that combining work and family is a good idea — suggesting they have acquired new information on the issue.
To the point that women anticipate longer careers when they decide how much education to acquire, the researchers combined many different datasets: some survey young people, some survey older people, and some are panels that interview the same people over time.
“We show that women who graduated high school by the late 1970s greatly overestimated the degree to which they would be homemakers, while those who graduated high school during and after the late 1970s, greatly underestimated it,” Washington said. “But how can women be miscalculating the impact of motherhood, an event for which there is so much data available in the form of friends and family we can ask and observe?”
Washington and her coauthors suggest that, in the past, motherhood was becoming easier through shifting attitudes about gender roles, new appliances that made housework less time consuming, and the availability of baby formula that reduced the need to breastfeed. In recent decades, motherhood has become perceived as more onerous as women feel pressured to breastfeed and to be closely involved in their children’s lives.
“The costs of motherhood had been decreasing, but now they’re increasing,” Washington said. “Simply asking your mother about her experience raising children will not give you accurate information. Catching flack for not baking homemade brownies for the school bake sale wasn’t an issue for my mother.”
The study on tax evasion looks at taxpayer behavior in highly partisan counties as elections pull them into and out of alignment with the political party that controls the White House.
Washington and her coauthors uncovered three types of evidence suggesting that people are less likely to evade their taxes when they have a positive view of the government. First, the self-employed paid more in income taxes when they had the same party affiliation as the president. There was no change in less-easily evaded third-party reported income over the same period, suggesting that economic conditions were not responsible for the increase in self-reported income. Next, the data showed there is less bunching at the income cutoff for the maximum earned income tax credit — an indication that less tax evasion is occurring — when a county’s majority political party controls the executive branch. Finally, the study found a decrease in computer-triggered IRS auditing when a county’s politics are aligned with the president’s political party, suggesting a decrease in questionable taxpayer behavior.
The findings could indicate that people do not fully understand how the government spends tax revenue, Washington said.
“One reason why people may be unhappy with the government is that they don’t know what the government actually does with their tax money,” she said. “For example, people often drastically inflate the amount of money that is spent on foreign aid. When the Kaiser Family Foundation informed poll residents of the actual amount spent on foreign aid, the fraction saying we were spending too much fell by half.”
She recommends undergraduates take a course in public finance, which is one of her favorite classes to teach.
“It’s a great class that everyone should take even if they don’t want to be an economist,” she said. “It’s important to know the government’s role in the economy and how things like Social Security, Medicare, Medicaid, and the tax system work. It is how you become a well-informed voter.”
Increasing diversity in economics
Washington earned a bachelor’s degree in public policy from Brown University. After graduation, she spent a year in Ecuador teaching English. She had planned to pursue a graduate degree in public policy but her undergraduate adviser urged her to choose a discipline, such as political science or economics, in order to broaden her employment options.
“I decided to be an economist because they are taken most seriously at the public policy table,” she said.
She earned her Ph.D. from the Massachusetts Institute of Technology in 2003. She was the only woman of color in her class.
“Lack of diversity is a huge problem in economics,” she said.
Washington co-chairs the American Economic Association’s Committee on the Status of Minority Groups in the Economics Profession, which seeks to diversify the profession at all levels. The group’s signature program is the Summer Training Program, currently at Michigan State University, which funds coursework in economic theory and research to prepare undergraduates to apply for economics Ph.D. programs. The group also sponsors a mentoring program for minority graduate students and runs a summer fellows program that places women and minority graduate students and early career postdocs in sponsoring research institutions. The committee sponsors the Div. E.Q. website, which offers advice and guidance for promoting inclusivity in economics classrooms and departments.
“We have to change many things, starting with how we introduce people to economics,” said Washington, who is the economics department’s director of undergraduate studies. “Students learn about widgets early on, but we don’t do enough to really show them what economists really spend their time thinking about. They have no idea that you can use economics to study the Voting Rights Act. They don’t realize the incredibly broad range of interesting questions that economics can address.”