Papers - Natural

John A List, Ian Muir, Devin Pope, Gregory Sun

Cited by*: None Downloads*: None

Left-digit bias (or 99-cent pricing) has been discussed extensively in economics, psychology, and marketing. Despite this, we show that the rideshare company, Lyft, was not using a 99-cent pricing strategy prior to our study. Based on observational data from over 600 million Lyft sessions followed by a field experiment conducted with 21 million Lyft passengers, we provide evidence of large discontinuities in demand at dollar values. Approximately half of the downward slope of the demand curve occurs discontinuously as the price of a ride drops below a dollar value (e.g. $14.00 to $13.99). If our short run estimates persist in the longer run, we calculate that Lyft could increase its profits by roughly $160M per year by employing a left-digit bias pricing strategy. Our results showcase the robustness of an important behavioral bias for a large, modern company and its persistence in a highly-competitive market.

John A List, Fatemeh Momeni

Cited by*: None Downloads*: None

We use a natural field experiment in which we hired over 2000 workers from an online labor market to explore how upfront payment affects worker motivation and misbehavior on the job. We start with a simple theory that shows paying upfront can increase misbehavior through reducing the perceived costs of cheating, but it can decrease misbehavior through generating a gift-exchange effect. Motivated by the theory, we designed a task that provided workers with opportunities to reciprocate or misbehave. A unique aspect of our design is that we are permitted an opportunity to measure the curvature of the gift-exchange value of the upfront payment. Our results suggest paying workers upfront induces a gift-exchange effect that is concave in the share of total wage paid upfront. Moreover, the impact is strong enough to suggest that small upfront payments are a cost-effective means for an employer to curb employee misbehavior.

Joshua D Angrist, Eric Bettinger, Michael Kremer

Cited by*: 2 Downloads*: 20

Colombia's PACES program provided over 125,000 poor children with vouchers that covered half the cost of private secondary school. The vouchers were renewable annually conditional on adequate academic progress. Since many vouchers were assigned by lottery, program effects can reliably be assessed by comparing lottery winners and losers. Estimates using administrative records suggest the PACES program increased secondary school completion rates by 15-20 percent. Correcting for the greater percentage of lottery winners taking college admissions tests, the program increased test scores by two-tenths of a standard deviation in the distribution of potential test scores. Boys, who have lower scores than girls in this population, show larger test score gains, especially in math.

Laura Derksen, Adamson Muula, Joep van Oosterhout

Cited by*: 0 Downloads*: 18

The HIV epidemic in southern Africa has important consequences for economic development. The epidemic could be stopped by a universal test and treat policy, as antiretroviral drugs block the spread of the virus. However, demand for HIV testing and treatment are surprisingly low. This paper develops a model in which the decision to seek an HIV test is a signal of infection, and those who seek a test are subject to statistical discrimination from potential sexual partners. We evaluate an information experiment designed to test the theory, and find evidence that this form of discrimination is a significant barrier to HIV testing. In particular, we provide information at the community level on the public benefit of antiretroviral therapy: because the drugs prevent HIV transmission, a person who is tested and treated for HIV is a relatively safe sexual partner. This information reduces discrimination and increases HIV testing, with the strongest effects in communities where the new information becomes common knowledge. The results demonstrate that discrimination towards HIV positive individuals can be due to rational behavior by a misinformed public, and that providing new information can be an effective way to mitigate its effects.

Paul W Rhode, Koleman S Strumpf

Cited by*: 12 Downloads*: 16

Political stock markets have a long history in the United States. Organized prediction markets for Presidential elections have operated on Wall Street (1880-1944), the Iowa Electronic Market (1988-present), and TradeSports (2001-present). Proponents claim such markets efficiently aggregate information and provide forecasts superior to polls. An important counterclaim is that such markets may be subject to manipulation by interested parties. We analyze this argument by studying alleged and actual speculative attacks- large trades, uninformed by fundamentals, intended to change prices- in these three markets. We first examine the historical Wall Street markets where political operatives from the contending parties actively and openly bet on city, state and national races; the record is rife with accusations that parties tried to boost their candidates through investments and wash bets. Next we report the results of a field experiment involving a series of planned, random investments-- accounting for two percent of total market volume-- in the Iowa Electronic Market in 2000. Finally, we investigate the speculative attacks on TradeSports market in 2004 when a single trader made a series of large investments in an apparent attempt to make one candidate appear stronger. In the cases studied here, the speculative attack initially moved prices, but these changes were quickly undone and prices returned close to their previous levels. We find little evidence that political stock markets can be systematically manipulated beyond short time periods.

Jan Hansen, Carsten Schmidt, Martin Strobel

Cited by*: 6 Downloads*: 13

Political stock markets (PSM) are sometimes seen as substitutes for opinion polls. On the bases of a behavioural model, specific preconditions were drawn out under which manipulation in PSM can weaken this argument. Evidence for manipulation is reported from the data of two separate PSM during the Berlin 1999 state elections.

Craig Gallet, John A List

Cited by*: 0 Downloads*: 1

This paper uses market share data to infer the nature of rivalry in the U.S. cigarette industry over the 1934-94 period. Unlike previous studies, which measure rivalry from various constructs of market share instability, we examine the time-series properties of market shares to determine whether or not rivalry is evident. Our empirical results imply that a majority of firm-level market shares are martingales, suggesting market shares have been unstable from 1934-94. This result leads us to conclude that rivalry in the cigarette industry has remained strong.

John A List, Daniel Rondeau

Cited by*: 31 Downloads*: 11

This study designs a natural field experiment linked to a controlled laboratory experiment to examine the effectiveness of matching gifts and challenge gifts, two popular strategies used to secure a portion of the $200 billion annually given to charities. We find evidence that challenge gifts positively influence contributions in the field, but matching gifts do not. Methodologically, we find important similarities and dissimilarities between behavior in the lab and the field. Overall, our results have clear implications for fundraisers and provide avenues for future empirical and theoretical work on charitable giving.

Richard Martin, John Randal

Cited by*: 0 Downloads*: 11

We describe a natural field experiment investigating donation behaviour. The setting was an art gallery where donations could be deposited into a transparent box in the foyer. Two aspects of the donation environment were manipulated: signs on the donation box and the initial contents of the box. We used three sign treatments: a control with no sign, a sign that thanked donors, and a sign that indicated donations would be matched. We used two initial contents treatments: one with relatively little money ($50) and one with four times as much. The average donation per donor was significantly larger in the $200 treatments but this was offset by a decrease in the propensity to donate. In the matching treatments donations were significantly larger both at the per donor and per visitor level. A control donate variable turned out to have the largest influence on donation behaviour: the day of the week. The average donation per visitor was 51% higher on Sundays, when compared to every other day of the week.

Uditi Karna, Min Sok Lee, John A. List, Andrew Simon, Haruka Uchida

Cited by*: Downloads*:

Educational disparities remain a key contributor to increasing social and wealth inequalities. To address this, researchers and policymakers have focused on average differences between racial groups or differences among students who are falling behind. This focus potentially leads to educational triage, diverting resources away from high-achieving students, including those from racial minorities. Here we focus on the "racial excellence gap" - the difference in the likelihood that students from racial minorities (Black and Hispanic) reach the highest levels of academic achievement compared with their non-minority (white and Asian) peers. There is a shortage of evidence that systematically measures the magnitude of the excellence gap and how it evolves. Using longitudinal, statewide, administrative data, we document eight facts regarding the excellence gap from third grade (typically ages 8-9) to high school (typically ages 14-18), link the stability of excellence gaps and student backgrounds, and assess the efficacy of public policies. We show that excellence gaps in maths and reading are evident by the third grade and grow slightly over time, especially for female students. About one third of the gap is explained by a student's socioeconomic status, and about one tenth is explained by the school environment. Top-achieving racial minority students are also less likely to persist in excellence as they progress through school. Moreover, state accountability policies that direct additional resources to reduce non-race-based inequality had minimal effects on the racial excellence gaps. Documenting these patterns is an important step towards eliminating excellence gaps and removing the "racial glass ceiling".

John A List, Warren McHone

Cited by*: 0 Downloads*: 9

In this article, we use annual (1980-90) county-level manufacturing plant location data for New York State to examine the effects of the 1977 Clean Air Act Amendments on the location decisions of new pollution-intensive manufacturing plants in the "neo-regulatory" (1980-84) and "mature-regulatory" (1985-90) phases of the Act's implementation. Our results suggest that the temporal effects of regulation vary. Whereas the location decisions of pollution intensive manufacturing firms were unaffected by the Act's regulatory restrictions in the "neo-regulatory" period, the restrictions appear to have had a significant negative impact on the location decisions of these types of firms in the Act's "mature-regulatory" phase. The diversion of new pollution intensive plants to counties with less stringent environmental regulations suggests that current US environmental regulations may be leading to a "browning process" whereby counties historically free of pollution become havens for polluters.

Katherine L. Milkman, Sean F. Ellis, Dena M. Gromet, Youngwoo Jung, Alex S. Luscher, Rayyan S. Mobarak, Madeline K. Paxson, Ramon A. Silvera Zumaran, Robert Kuan, Ron Berman, Neil A. Lewis Jr, John A. List, Mitesh S. Patel, Christophe Van den Bulte, Kevin G. Volpp, Maryann V. Beauvais, Jonathon K. Bellows, Cheryl A. Marandola, Angela L. Duckworth

Cited by*: None Downloads*: None

Encouraging routine COVID-19 vaccinations will probably be a crucial policy challenge for decades to come. To avert hundreds of thousands of unnecessary hospitalizations and deaths, adoption will need to be higher than it was in the autumn of 2022 or 2023, when less than one-fifth of Americans received booster vaccines. One approach to encourage vaccination is to eliminate the friction of transportation hurdles. Previous research has shown that friction can hinder follow-through and that individuals who live farther from COVID-19 vaccination sites are less likely to get vaccinated. However, the value of providing free round-trip transportation to vaccination sites is unknown. Here we show that offering people free round-trip Lyft rides to pharmacies has no benefit over and above sending them behaviourally informed text messages reminding them to get vaccinated. We determined this by running a megastudy with millions of CVS Pharmacy customers in the United States testing two effects: free round-trip Lyft rides to CVS Pharmacies for vaccination appointments, and seven different sets of behaviourally informed vaccine reminder messages. Our results suggest that offering previously vaccinated individuals free rides to vaccination sites is not a good investment in the United States, which is contrary to the high expectations of both expert and lay forecasters. Instead, people in the United States should be sent behaviourally informed COVID-19 vaccination reminders, which increased the 30-day COVID-19 booster uptake by 21% (1.05 percentage points) and spilled over to increase 30-day influenza vaccinations by 8% (0.34 percentage points) in our megastudy. More rigorous testing of interventions to promote vaccination is needed to ensure that evidence-based solutions are deployed widely and that ineffective but intuitively appealing tools are discontinued.

Jie Bai

Cited by*: 0 Downloads*: 157

There is often a lack of reliable high quality provision in many markets in developing countries. I designed an experiment to understand this phenomenon in a setting that features typical market conditions in a developing country: the retail watermelon market in a major Chinese city. I begin by demonstrating empirically that there is substantial asymmetric information between sellers and buyers on sweetness, the key indicator of quality for watermelons, yet sellers do not sort and price watermelons by quality. I then randomly introduce one of two branding technologies into 40 out of 60 markets-one sticker label that is widely used and often counterfeited and one novel laser-cut label. I track sellers' quality, pricing and sales over an entire season and collect household panel purchasing data to examine the demand side's response. I find that laser branding induced sellers to provide higher quality and led to higher sales profits, establishing that reputational incentives are present and can be made to pay. However, after the intervention was withdrawn, all markets reverted back to baseline. To rationalize the experimental findings, I build an empirical model of consumer learning and seller reputation. The structural estimates suggest that consumers are hesitant to upgrade their perception about quality under the existing branding technology, which makes reputation building a low return investment. While the new technology enhances consumer learning, the resulting increase in profits is not sufficient to cover the fixed cost of the technology for small individual sellers. Counterfactual analysis shows that information frictions and fragmented markets lead to significant under-provision of quality. Third-party interventions that subsidize initial reputation building for sellers could improve welfare.

John Gibson, David McKenzie, Steven Stillman

Cited by*: 3 Downloads*: 13

People migrate to improve their well-being, whether through an expansion of economic and social opportunities or a reduction in persecution. Yet a large literature suggests that migration can be a very stressful process, with potentially negative impacts on mental health reducing the net benefits of migration. However, to truly understand the effect of migration on mental health one must compare the mental health of migrants to what their mental health would have been had they stayed in their home country. The existing literature is not able to do this and typically settles for comparing the mental health of migrants to that of natives in the destination country, which takes no account of any pre-existing differences between these groups. This paper overcomes the selection problems affecting previous studies of the effect of migration on mental health by examining a migrant lottery program. New Zealand allows a quota of Tongans to immigrate each year with a lottery used to choose amongst the excess number of applicants. A unique survey conducted by the authors in these two countries allows experimental estimates of the mental health effects of migration to be obtained by comparing the mental health of migrants who were successful applicants in the lottery to the mental health of those who applied to migrate under the quota, but whose names were not drawn in the lottery. Migration is found to lead to improvements in mental health, particularly for women and those with poor mental health in their home country.

Guodong Gao, Tianshu Sun, Ginger Zhe Jin

Cited by*: 0 Downloads*: 64

While much research has examined the role of technology in moderating online user connections, how IT motivates offline interactions among users is much less understood. Using a randomized field experiment involving 80,000 participants, we study how mobile messaging can leverage recipients' social ties to encourage blood donation. There are three main findings: first, both behavior intervention (in the form of reminder message) and economic reward (in the form of individual or group reward) increase donations, but only the messages with group reward are effective in motivating more donors to donate with their friend(s); second, group reward tends to attract different types of donors, especially those who are traditionally less active in online social setting; and third, across all treatments, message recipients donate a greater amount of blood if their friends are present. Structural estimation further suggests that rewarding group donors is four times more cost-effective than rewarding individual donors. Based on the structural estimates, we perform policy simulations on the optimal design of mobile messaging. The method of combining structural model and randomized field experiment opens new frontiers for research on leveraging IT to mobilize a user's social network for social good.

Ryan D Friedrichs

Cited by*: 0 Downloads*: 14

No abstract available

Donald P Green

Cited by*: 0 Downloads*: 1

This essay summarizes the results of a large-scale randomized experiment conducted during the 2000 election campaign by the NAACP National Voter Fund, which sought to mobilize African-American voters. Focusing solely on the direct mail and phone banking components of the NAACP-NVF campaign, this study examines the voting behavior of 980,208 participants residing in single-voter households, 1.7 percent of whom were randomly assigned to a control group. The experiment permits us to estimate (1) the extent to which the National Voter Fund's phone calls and direct mail increased voter turnout and (2) the approximate cost per vote. Within this sample, the NVF's two pieces of GOTV mail, three live phone calls, and two recorded phone calls had modest effects, generating approximately 7,100 additional votes at $158 per vote. The upper bound of a 90 percent confidence interval puts these figures at 16,214 additional votes at $69 per vote.

Melissa R Michelson, Herbert Villa Jr.

Cited by*: 0 Downloads*: 21

Less than a third of Latinos vote in Presidential elections, while less than one fourth participate in Congressional elections. Turnout among young Latinos (age 18-25) is even lower. This paper describes the results of a field experiment aimed at increasing turnout among young Latinos in Fresno, California conducted in the fall of 2002. Canvassers went door-to-door during the final two weekends before Election Day to urge registered young people to go to the polls. Young people of all races/ethnicities were targeted. In addition to testing the effectiveness of personal contact and how this varies among registered voters of various races/ ethnicities, the project also included two imbedded experiments. First, the race/ethnicity of the canvassers was randomly assigned, to test whether Latinos and non-Latinos are equally effective at getting Latinos and non- Latinos to the polls. Second, the message delivered to contacted registered voters was randomly assigned, to test whether young Latinos are more receptive to a message which stresses group solidarity or one that emphasizes civic duty. The experiment demonstrates that Latino canvassers are better than non-Latinos at contacting young Latino voters, and that young Latinos are more receptive than are non-Latinos to door-todoor mobilization efforts.

Ryan D Friedrichs, David C King, David W Nickerson

Cited by*: 0 Downloads*: 5

Recent large-scale field experiments of get out the vote (GOTV) drives have been non-partisan and may not accurately capture the effectiveness of partisan campaign outreach. In the 2002 Michigan gubernatorial election, a large field experiment across 14 state house districts evaluated the cost effectiveness of three mobilization technologies utilized by the Michigan Democratic Party's Youth Coordinated Campaign: door hangers, volunteer phone calls, and face-to-face visits. The results indicate that all three GOTV strategies possess similar cost-effectiveness.

Benjamin A Olken

Cited by*: 19 Downloads*: 38

This paper uses a randomized field experiment to examine several approaches to reducing corruption. I measure missing expenditures in over 600 village road projects in Indonesia by having engineers independently estimate the prices and quantities of all inputs used in each road, and then comparing these estimates to villages' official expenditure reports. I find that announcing an increased probability of a government audit, from a baseline of 4 percent to 100 percent, reduced missing expenditures by about 8 percentage points, more than enough to make these audits cost-effective. By contrast, I find that increasing grass-roots participation in the monitoring process only reduced missing wages, with no effect on missing materials expenditures. Since materials account for three-quarters of total expenditures, increasing grass-roots participation had little impact overall. The findings suggest that grass-roots monitoring may be subject to free-rider problems. Overall, the results suggest that traditional top-down monitoring can play an important role in reducing corruption, even in a highly corrupt environment.