Saturday, June 1
Behavioral Gender
9:00 - 11:00 am
While there is a vast (and mixed) literature on gender differences in social preferences, little is known about believed gender differences in social preferences. Using data from 15 studies and 8,979 individuals, we find that women are believed to be more generous and more equality-oriented than men. This believed gender gap is robust across a wide range of contexts that vary in terms of strategic considerations, selfish motives, fairness concepts, and payoffs. Yet, this believed gender gap is largely inaccurate. Consistent with models of associative memory, and specifically the role of similarity and interference, the believed gender gap is correlated with recalled prior life experiences from similar contexts and significantly affected by an experience that may interfere with the recall process of prior memories even though this interfering experience should not affect the beliefs of perfect-memory Bayesians. Application studies further reveal that believed gender differences extend to the household (i.e., beliefs about contributions to the home, family, and upbringing of children), the workplace (i.e., beliefs about equal pay) and policy views (i.e., beliefs about redistribution, equal access to education, healthcare, and affordable housing).
Rania Gihleb - University of Pittsburgh - "Gender-specific labor market shocks and household bargaining power"
Using machine learning and product-level data describing single men's and women's consumption patterns, we create an index that quantifies the ``gendered'' nature of consumer goods. We use the index to investigate the impact of gender-specific shocks to spouses' economic stature on the budget allocation within heterosexual married households toward products predominantly favored by either gender. Specifically, our findings indicate that the adoption of robots, which heightened women's economic standing relative to men, led to a noticeable shift in consumption patterns, favoring products more frequently bought by single women. Conversely, the expansion of fracking, which bolstered the demand for young, less-skilled men, resulted in an increased consumption of products predominantly preferred by men. We then extend our analysis to children's products. While neither of the shocks affected the expenditure allocated to children's products, positive shocks to the economic stature of women led to an increase in expenditure on products prevalently bought for daughters (rather than sons).
Amanda Agan - Rutgers University - "Automating Automaticity: How the Context of Human Choice Affects the Extent of Algorithmic Bias"
Consumer choices are increasingly mediated by algorithms, which use data on those past choices to infer consumer preferences and then curate future choice sets. Behavioral economics suggests one reason these algorithms so often fail: choices can systematically deviate from preferences. For example, research shows that prejudice can arise not just from preferences and beliefs, but also from the context in which people choose. When people behave automatically, biases creep in; snap decisions are typically more prejudiced than slow, deliberate ones, and can lead to behaviors that users themselves do not consciously want or intend. As a result, algorithms trained on automatic behaviors can misunderstand the prejudice of users: the more automatic the behavior, the greater the error. We empirically test these ideas in a lab experiment, and find that more automatic behavior does indeed seem to lead to more biased algorithms. We then explore the large-scale consequences of this idea by carrying out algorithmic audits of Facebook in its two biggest markets, the US and India, focusing on two algorithms that differ in how users engage with them: News Feed (people interact with friends’ posts fairly automatically) and People You May Know (people choose friends fairly deliberately). We find significant out-group bias in the News Feed algorithm (e.g., whites are less likely to be shown Black friends’ posts, and Muslims less likely to be shown Hindu friends’ posts), but no detectable bias in the PYMK algorithm. Together, these results suggest a need to rethink how large-scale algorithms use data on human behavior, especially in online contexts where so much of the measured behavior might be quite automatic.
We study the extent to which delaying pregnancy mitigates the impact of children on women’s labor market outcomes. We leverage quasi-random variation in the timing of pregnancy in a setting where women intend to delay having children by using long-acting reversible contraceptives. While most women successfully delay pregnancy, some have unplanned pregnancies. Analyzing linked health and labor market data from Sweden, we find that unplanned pregnancies halt women’s career progression resulting in income losses of 20% by five years after the unplanned pregnancy. Using pregnancy as an instrument for birth in a dynamic treatment effect framework, the detrimental effects of unplanned children are larger for younger women and women enrolled in education. This indicates that unplanned first births are particularly disruptive early on when women are investing in their human capital. In contrast, we find smaller impacts of unplanned pregnancies for women who are already mothers. When we estimate the impact of first births identified from quasi-random success of fertilization procedures, we likewise find smaller impacts of children among women who intend to get pregnant.
Behavioral Health
11:20 am - 12:50 pm
Discrimination against doctors is important but scantly studied. I report a field experiment which observes that customers discriminate against Black and Asian doctors when they choose healthcare providers, and that this can be substantially reduced by supplying information on physician quality. I evaluate customer preferences in the field with an online platform where cash-paying consumers can shop and book a provider for medical procedures based on a novel experimental paradigm. Actual paying customers evaluate doctor options they know to be hypothetical to be matched with a customized menu of real doctors, preserving incentives. Racial discrimination reduces patient willingness-to-pay for Black and Asian doctors by 12.7% and 8.7% of the average colonoscopy price respectively; customers are willing to travel 100-250 miles to see a white doctor instead of a Black doctor, and somewhere between 50-100 to 100-250 miles to see a white doctor instead of an Asian doctor. Providing signals of doctor quality reduces this willingness-to-pay racial gap by about 90%. Willingness-to-pay penalties on minority doctors are multiples of actual average racial quality differences and even the difference between doctors in highest and lowest quality levels. This field evidence shifts the focus beyond traditional taste-based and statistical discrimination to include behavioral mechanisms like biased beliefs and deniable prejudice. Discrimination against Black doctors are higher for non-college-graduate customers and residents in zipcodes that voted for the 2020 presidential candidate on the political right. Actual booking behavior allows cross-validation of incentive compatibility of the stated preference elicitation.
We hypothesize that deepening resource scarcity will result in rationing on the basis of group identity in settings with underlying discrimination. We provide evidence of such consequential race-based rationing in a high-stakes setting: health care. Using detailed, time-stamped data on 107,000 patient admissions to a large health system, we find that in-hospital mortality increased for Black, but not White, patients as hospitals reached capacity (a state of resource scarcity where biases in decision-making are likely to emerge or be exacerbated). We identify rationing by wait times as a mechanism, while documenting that sick Black patients waited longer for care than healthy White patients at every capacity level, likely due to systematic misassessment of medical need. Text analysis techniques on unstructured provider notes reveal differential rationing of provider effort by race as another potential mechanism. Together, these findings demonstrate important linkages between three key economic concepts: scarcity, discrimination, and rationing.
Mario Macis - Johns Hopkins University - “How Do Surrogates Make Treatment Decisions for Patients with Dementia: An Experimental Survey Study”
Despite the growing need for surrogate decision-making for older adults, little is known about how surrogates make decisions and whether advance directives would change decision-making. We designed an experimental, vignette-based survey that cross-randomized cognitive impairment, gender, the relationship between surrogate and patient, presence and characteristics of advance care planning, framing of the decision, the patient doctor's recommendation, and prevailing social norms at the patient's hospital. We administered the survey to three distinct cohorts from the nationally representative Ipsos Knowledge Panel, each comprising approximately 2,000 US adults aged 18 and above. The first group was a random selection, the second consisted of caregivers, and the third included individuals with chronic illnesses. We have five key results: 1) Respondents were much less likely to recommend life-sustaining treatments for patients with dementia; 2) Respondents frequently ignore patient preferences for life-extending treatment, especially when the patient has dementia, and were more likely to choose unwanted life-extending treatments for patients without dementia; 3) The "dementia penalty" was somewhat larger among respondents who are chronically ill and, especially, respondents who are caregivers for another adult; 4) Both doctors' recommendations and prevailing social norms influence the respondents' choices; and 5) Respondents were more likely to choose treatments that matched their own preferences when there was uncertainty about the patient's wishes. These findings underscore the need for improved communication and decision-making processes for patients with cognitive impairment and highlight the importance of choosing a surrogate decision-maker with similar treatment preferences.
Behavioral Development
1:50 - 3:20 pm
Gabriel Tourek - University of Pittsburgh - "Preferences and Performance Among Bureaucrats: Evidence from Pakistan's Civil Service"
Does assignment to more preferred postings improve bureaucrats’ performance? We assemble administrative data on bureaucrats within Pakistan’s civil service between 2008 and 2021. We exploit variation in allocation to postings at service entry arising from sharp discontinuities in the likelihood of assignment to more preferred posts. A stronger match between preferences and assignment increases performance during pre-deployment training. Surveys with bureaucrats suggest roles for intrinsic motivation, career and prestige concerns. We are in the process of examining impacts on longer-run career and performance outcomes.
Reshmaan Hussam - Harvard University - "The Household at Work: A Field Experiment in the Rohingya Refugee Camps"
We use a randomized control trial to assess how employment affects the wellbeing of the household -- as distinct from effects on the individual supplying the labor. We work with 2513 households in the world's largest refugee camp, and randomly assign the husband or wife to an employment intervention. We also include cash and unpaid work treatment arms in order to disentangle underlying channels driving the employment effects we observe. While gainful employment delivers comparable psychosocial benefits to both men and women, spillovers are asymmetric: households in which men are treated fare significantly better as measured by psychosocial wellbeing and domestic violence for both partners. Our results are consistent with individuals preferring that their families inhabit traditional gender norms and motivate caution in the distribution and targeting of social protection programs.
Anne Karing - University of Chicago - "Incentives and Motivation Crowd-Out: Experimental Evidence from Childhood Vaccination"
We examine the impact of incentives on intrinsic motivation after their removal. We follow up with parents three years after exposure to a bracelet incentive given to children upon timely vaccination in Sierra Leone. We leverage the design of an experiment under which clinics were randomly assigned to either offer incentives or not. Since only parents who had a newborn at the time were eligible, we also exploit individual variation in exposure within clinics. First, we find that eligibility to an incentive for an earlier child reduces parents' motivation to vaccinate their subsequent child on time, with decreases of 5 to 11 percent in the number of timely visits compared to unexposed parents. There are no impacts on vaccination rates by 15 months of age, suggesting parents delay vaccination, but do not abstain. Second, parents living in incentive communities who were ineligible at the time are unaffected, ruling out that results are driven by changes in community norms or clinic practices. Third, incentives that focus parents' attention on caring for their child's health do not lead to adverse effects. Together, our results suggest that incentives that shift parents' attention to an external reward, can crowd-out intrinsic motivation and negatively impact behaviors once removed.
Behavioral Labor
3:40 - 5:10 pm
Marta Serra-Garcia - University of California, San Diego
This paper uses linked survey-administrative data to examine workers’ beliefs about firm wages and rents. We document that workers expect their pay would vary across firms, even within narrowly defined labor markets and that perceived premia are correlated with observed pay policies. Firm-specific hypothetical choice experiments reveal that workers also expect firms to vary in the rents they offer, and that they expect high wage firms to offer more rents. We document that workers use these beliefs to direct their search.
Zoe Cullen - Harvard University
Sunday, June 2
Behavioral Development
9:00 - 10:30 am
Joshua Dean - University of Chicago - "Demand for Urban Exploration: Evidence from Nairobi"
Simone Schaner - University of Southern California - "Better Together? The Impact of Cash Transfers and Couples' Financial Planning on Household Dynamics and Economic Outcomes”
The receipt of large cash transfers by poor households creates a high-stakes resource allocation decision that has substantial economic implications and may lead to intrahousehold conflict. We conducted a randomized trial in rural Liberia, analyzing the effects of a large unconditional cash transfer implemented with and without a light-touch facilitated financial planning exercise. This add on intervention was designed to provide couples with an opportunity to jointly plan their transfer spending before disbursement, and offered soft commitment tools to help couples adhere to their plans. We find large positive effects of the cash transfer on a wide range of economic outcomes; and the couples' planning intervention significantly amplified these positive effects on income, food security, and housing quality. We find no evidence that planning helped couples stick to baseline preferences; rather it appears planning may have helped couples identify better uses for new resources. However, we also find evidence that planning-induced income gains are sometimes accompanied by non-economic costs. Counter to the design intent behind planning, “conflict vulnerable” couples exposed to planning experienced both increased income and elevated rates of intimate partner violence.
Muhammad Yasir Khan - University of Pittsburgh
Behavioral Labor
10:50 am - 12:20 pm
People and organizations often set goals to self-motivate and achieve better outcomes in challenging tasks. But self-set goals, and their effectiveness, might depend on what people expect to happen. Do goals reflect expectations or do goals set expectations? How do goals and expectations affect performance? We run an online real-effort task to answer these two questions by introducing exogenous variation in goals and expectations. First, we find that goals mostly reflect expectations rather than affect them. Second, practicing an easier version of a task leads to higher expectations and higher performance. Eliciting a goal leads to higher performance as well. However, controlling for expectations, changing the difficulty of the goal has no discernible effect. These results suggest that people should come in optimistic and set a goal, but they cannot fool themselves into expecting and doing more simply by choosing a higher goal.
Björn Bartling - University of Zurich - "Paternalistic Interventions: Determinants of Demand and Supply"
Individuals sometimes engage in behaviors that negatively impact their well-being, creating opportunities for paternalistic interventions by governments, experts, or caregivers. This experimental study investigates both the circumstances under which individuals decide to intervene in the welfare of others, and when they are inclined to seek such interventions for themselves. Utilizing a general sample of the U.S. population, we study how freedom of choice, consent rights, confidence, and trust affect attitudes towards paternalistic interventions. A better understanding of these attitudes can contribute to the design of effective and widely accepted policies across multiple domains, including health policy.
Alain Cohn - University of Michigan - "Social Media and Job Market Success: A Field Experiment on Twitter"
We conducted a field experiment on Twitter to examine the impact of social media promotion on job market outcomes in economics. Half of the 519 job market papers tweeted from our research account were randomly assigned to be quote-tweeted by prominent economists. Quote-tweeted papers received 442% more views and 303% more likes. Moreover, candidates in the treatment group received one additional flyout, with women receiving 0.9 more job offers. These findings suggest that social media promotion can improve the visibility and success of job market candidates, especially for underrepresented groups such as women, promoting diversity and inclusion in economics.