Updated 8/22/2021 to display Fall 2021 projects
We believe that conducting research is an important part of your training as an economist. Learning how to define a question, locate and review related literature, identify, download and clean data, and conduct empirical analysis develops a skill set highly valued by graduate programs and future employers alike. The Economics Department has two programs:  the general undergraduate student researcher program, and  the Lynde grad student mentoring program. The mentoring program is a subset of the more general program. You sign up only once. . . Read on.
The undergrad student researcher program is designed for Economics majors at any stage of your education. If you have not yet taken econometrics, we still encourage you to apply. Some graduate students or faculty need RAs who can do coding in Stata or R or Python, but others are looking for RAs to do literature searches or other tasks that do not require econometrics or CS/DS toolsets.
The grad student mentoring program adds in several components that ensure the undergraduate student researcher experience provides you not only with an experience to list on your resume, but also with a graduate student mentor who will teach you about research and who is available to discuss courses, work or graduate school plans, the life of an economist, and whatever else may come up. The mentoring program is funded by a generous donation from Cal alumnus, Dr. Matthew Lynde.
About the Lynde Mentoring Program
The features of the mentoring program are listed below.
- You work with a graduate student who has expressed interest in and received training in serving as a mentor for undergraduates. Mentoring means that the graduate student has made a commitment to discussing the research process with the RA, helping the RA see the big picture for the research project and the specific tasks the RA is working on, talking with the RA about future opportunities in economics or in research (including graduate school options), and ensuring that the RA acquires research skills as part of the RA process. If you are considering graduate school in economics, or just want to know more about what it means to be an economist, the mentoring program may be for you.
- A welcoming reception for all RAs hired as part of the Mentoring Program will be held on Tuesday evening, September 14, 6:30-8:00. The location is still to be determined. The welcoming reception will give you the opportunity to meet all the other undergraduate RAs serving this term, learn about all of this term's projects, and ask questions about research in economics.
- A pair of zoom-workshops conducted by Economics graduate student will offer you tips & training in the use of R and LaTeX. The workshops will be scheduled for the 4th or 5th week of the term. More information will be on the RSVP form for the welcoming reception.
- An end-of-term Zoom-poster reception scheduled for Thursday evening December 2, 6:30 - 8:00 pm, in 611 Evans Hall (Peixotto Room), at which you and the grad student you're working with will display a poster that describes the project and the work you've contributed to the project. Posters will then be put on display in the 6th floor of Evans Hall for at least a semester.
- A small financial incentive ($50) will be paid in December to undergrad RAs who participate in the mentoring program by working with a grad student mentor, attending the welcoming reception on September 14 and attending the poster session on December 2.
There are about 40 graduate students who have completed the training to serve as mentors. Their projects cover a wide range of fields and the research tasks they are looking for help with also cover a wide range. The second column of the table below indicates whether or not the grad student is part of the mentoring program. (In one case, the project is listed as a faculty project but it is the grad student GSR who will be working with you.) If you are interested in participating in the mentoring program, be sure to check the appropriate box at the bottom of the application for serving as an undergrad RA.
There are also Economics, ARE, and BPP (part of Haas's PhD program) grad students who are not part of the mentoring program, postdocs, and ARE & Economics faculty who will look at the applications in order to find RAs. So you can be an RA without being in the mentoring program.
About the Application
The application form for undergraduate RA positions is available at https://www.econ.berkeley.edu/undergrad/economics-majors-interested-ra-positions. The Economics faculty and graduate students seeking RA assistance will reach out to those students they are interested in interviewing. Watch your email!
On the application form, you can signal up to three (3) projects that particularly interest you. Use the project number in the left most column when completing the "signalling" part of the application. The second column tells you whether or not the person in charge of the project is a faculty member or a graduate student who is part of the Mentoring Program.
About Earning Econ 199 Credits
Two important points about compensation:  You must be an enrolled student. Due to labor laws, we cannot allow people who have already graduated to serve as an unpaid RA.  Being an RA for a graduate student or faculty member is considered an educational or academic endeavor. In most cases, these are unpaid positions, much like an academic internship. If you are not being paid, you must receive 1-3 Econ 199 units for being an RA. These will be P/NP units that count toward graduation but not as an upper-division elective. Because the deadline to sign up for Econ 199 is Friday of the 3rd week of classes, all unpaid positions must be filled by then.
(Because of the cost of enrolling in summer units, Econ 199 enrollment not required in summer term.)
Non-Economics Faculty & Graduate Students Seeking RA Assistance
Graduate students or faculty from other departments can also hire our Economics undergraduates! We invite grad students, post-docs, and faculty who are not in the Economics Department to submit a description of their position directly to the undergraduate office (email@example.com) for inclusion in the weekly bCourses e-blast that goes to all Econ majors. Please include the following information: who you are, contact information, deadline for applying; title & description of the project; expected RA tasks and anticipated RA skill set; whether compensation is $ or units; anticipated # of hours per week. Send your info by email to firstname.lastname@example.org with an indication that it is for the weekly e-blast to Economics majors.
|Project #||In Mentoring Program?||Grad Student or Faculty||# of RA's needed||Project Title||Project Abstract||RA tasks|
|1||yes||Econ grad student mentor||3 new||Rational Inattention, Networks and the Propagation of Macroeconomic Shocks||This project aims at incorporating insights from Behavioral Economics, namely models of bounded rationality such as Rational Inattention, to study the propagation of macroeconomic shocks in a networked economy. In standard models, agents are fully aware of industry-specific shocks in any sector and geographical location in the economy, and know how to respond to them optimally. We relax this assumption and allow agents to pay less attention to shocks to firms that are far from them, either in a sectoral or a geographical sense, and analyze the consequence to the propagation of shocks in the macroeconomy.||
- Work with data used in estimation/calibration of the model
- Help with the numerical implementation of the model and running simulations
- Knowledge of one of the following computational packages required: R, Python, Matlab or Julia
|2||yes||Econ grad student mentor||2 new||Local Information and Inflation Expectations Formation||This project aims at studying the role of local information in the formation of inflation expectations. Recent research on the slope of the Phillips curve has highlighted the importance of inflation expectations to the dynamics of inflation. However, the exact process of how these expectations are formed is still largely an open question. In a networked economy setting, we investigate the contribution of local information to inflation expectation updating by studying the response to shocks at a disaggregated level. As an intermediate step, we construct novel data on local shocks by using methods such as figure to text classification and natural language processing.||
- Using Python to scrape, process and summarize data from scanned documents
- Help with Data analysis, model estimation and calibration
|3||yes||Econ grad student mentor||2 new||Cohabitant's effect on retirement decisions||Given that retirement benefits and medicaid comprise a majority of the entitlement programs provided through fiscal policy, it is important to understand how having an adult cohabitant affects these programs. Dependence on parents' income may also explain the low labor force participation experienced from the male labor force. To what extent should the entitlement programs be constrained to prevent "moochers" from forgoing participation in the labor force? I explore these questions using Health and Retirement Surveys.||
Literature review, summary statistics reproduction through the use of STATA
|4||yes||Econ grad student mentor||2 new||Detriments of being a tied spouse: Climbing the occupational ladder is more costly for military wives than non-military spouses.||Does your choice of partner matter for wage differentials? I test this question by documenting the upward occupational mobility of military wives compared to non-military spouses. Evidence shows that due to constant migration, military wives have lower labor force participation rates and lower earnings compared to non-military wife counterparts. I further explore this by analysing the occupations that military wives attain post migration. If military wives always acquire entry level positions rather than ever climbing the occupational ladder this may shed more light on the wage differentials among non-military wives and military wives. Furthermore, this question will determine if your choice of partner affects the wage differentials observed between men and women.||
Literature Review, Producing Basic Summary Statistics in STATA, Producing Graphs in STATA
|5||yes||Econ grad student mentor + Faculty (Supreet Kaur)||2 new||Promoting regular labor supply among the urban poor||High labor turnover and absenteeism are major impediments to productivity in poor countries We hypothesize that workers have difficulty providing regular labor supply in the formal sector because they lack the habit of doing so. We design an RCT with urban casual laborers in India, harnessing the predictions of a habit formation model. Laborers are provided incentives to boost their labor supply to urban labor stands -- their primary source of regular employment -- over a 2-month period. We examine the persistence of effects on labor supply once incentives are removed. Pilot results provide strong preliminary support for our hypothesis.||
Perform data cleaning and analysis, production of tables and figures, literature review.
|6||yes||Haas grad student mentor||3 new||From Rivals to Partners: The Alignment of Capital and State Coercion in the Rise of Modern Economic Growth||The conventional wisdom on the source of economic growth emphasizes inclusive institutions: constraints on state elites, which allow open access to political and economic power. Yet, in many contexts where economic growth has emerged, we see the partnership of powerful, coercive (often nondemocratic) states and private enterprise: from Industrial Revolution Britain, to the post-WWII Asian economic tigers, to post-Mao China. Focusing on the emergence of modern growth in Britain, we propose a model of a partnership between coercive states and private actors: merchants lend money to the state, which is used to develop coercive capacity. In our model, good institutions are an outcome, rather than a cause of increased economic activity, and the good outcome is driven by aligned incentives, not political inclusion. To study the openness of the British political elite in the 17th and 18th centuries, we collect data on the social origins of high officials, MP's, and Privy Councilors. We also collect information from the Canterbury Probate Records to examine the composition of the economic elite. Our preliminary results suggest an alignment between traditional political elites and new economic elites, which we hope to examine further using data on financial interests of the political elite, and on the relationship between international trade, wars, and government revenues.||
Our goal is to match RAs to tasks based on interest and skills. We expect RAs to mainly contribute to the empirical and data-intensive aspects of the project. We use Python, R and Stata. Tasks include, for example:
1) Collecting and digitizing novel archival data
2) Web scraping and Optical Character Recognition of historical records
3) Combining data from multiple sources to construct new data sets
4) Working with “text as data” using machine learning and natural language processing
5) Analyzing the data using econometric methods
6) Visualizing the results in graphs and maps
|7||yes||Haas grad student mentor||3 new||Organized Redistribution: Can Political Machines Promote Social Mobility? Evidence from Tammany Hall||Political machines are hierarchical organizations with a dense network of local brokers that mobilize votes for their candidates with promises of patronage, pork, or other rewards and threats. Machines dominate elections in many developing countries and are commonly associated with corruption and bad governance. Existing research emphasizes how brokers often target poorer voters. By targeting patronage and public services at poorer citizens, do machine politicians contribute to more equal outcomes and greater social mobility? Can disenfranchised groups like immigrants improve their situation by rising through the machine hierarchy? I study these questions in the context of the archetypical machine in US history: New York City’s Tammany Hall. I am currently digitizing archival records on the machine’s personnel, yearly registers of city employees, and local public good provision. I plan to link this information to individual-level census data and newly collected data on house prices to document who profits from machine politics. I would then compare the economic outcomes of immigrants living close to local machine operatives vs. those living in the same district but further away from operatives. To overcome potential selection, I plan to estimate a difference-in-difference specification and leverage the variation in the proximity to machine officials induced by unexpected changes in the machine’s personnel.||
The goal is to match RAs to tasks based on interest and skills. I expect RAs to mainly contribute to the empirical and data-intensive aspects of the project. We use Python, R, Stata, and GIS applications like QGIS or ArcGIS.
I am especially looking for someone with GIS experience to create new shapefiles of district boundaries for historical Manhattan. But these skills can also be learned while working on the project.
Other tasks include, for example:
1) Collecting and digitizing novel historical data
2) Web scraping and Optical Character Recognition of historical records
3) Working with geo-spatial data and digitized maps
4) Combining data from the census and other sources to construct new data sets
5) Analyzing the data using econometric methods
6) Visualizing the results in graphs and maps
|8||yes||Econ grad student mentor||1 new||Political Slant in Macroeconomic Research||We will be using natural language processing tools to quantify political views and ideology implicit in macroeconomic research.||
Natural language processing
|9||Econ faculty||4 new||History of Women Faculty in Economics||As part of the campus 150W project (https://150w.berkeley.edu/) marking the 150th anniversary of the admission of women to UCB, we have been & will be working on developing a history of women faculty in the Economics Department. See https://www.econ.berkeley.edu/women-history. The project for 2021-22 is to interview another 3-4 women who served on UCB Econ faculty in the post-1970s. This will wrap up the Econ Department 150W project.||
Literature search, research to prepare interview questions, conducting oral interviews of women faculty, writing and editing.
|10||yes||Econ grad student mentor||3 new||Unpacking Intergenerational (Im-)mobility: Child vs. Parent Career Preferences||With both students and parents having preferences over students' career choices, are students willing to adjust in case these diverge? If so, how does this affect intergenerational mobility? I design a field experiment with 1,200 students and 800 parents in Germany to test whether the public nature of actual decisions causes students to adjust to parents: I experimentally vary whether students' incentivized career aspirations are shared with their parents or not. Making students' aspirations observable by parents causes an increased share of students with at least one college-educated parent to state an aspiration to attend college and more students to aspire to high earning fields. As a result, the socio-economic gap in college aspirations doubles to 27 percentage points and also widens for aspirations to enroll in a high earning major at university. Estimating a model where career tracks are chosen under uncertainty suggests that the socio-economic gap in observable college aspirations is in almost equal terms due to this adjustment to parents and differences in beliefs across backgrounds. Parents' career preferences and children's tendency to adjust to these can therefore contribute to intergenerational immobility.||
This semester I will mostly need help with drafting literature reviews and brief summaries (extracting highlights) of individual papers in a few literatures, for example:
- social mobility/occupational persistence in Sociology
- psychology literature on children's/adolescents' desire for parents' approval, fear to disagree with parents etc.
- pecuniary and non-pecuniary returns to college/majors [especially for those at the margin of going]
- there might be more topics that will generally be related to the overall research theme above
You should bring a passion for the general research topic outlined above, an interest in expanding your knowledge on one of the topics above and the ability to identify, skim and summarize relevant papers. You would not need to cover all of these topics and could focus on one of them. You do not need to be acquainted with these literatures yet, although if you have a sociology or psychology background, you're welcome to make use of it!
|11||yes||Econ faculty||1 new||Effects of minimum wages on small businesses and on children's wellbeing||This project has two parts: 1. It is often assumed that small businesses are especially vulnerable to minimum wage increases. We will test this assumption using Current Population Survey data since 1990 on low-wage workers and low-wage industries. Our methods employ state-of-the-art causal econometric models. 2. Do minimum wage increases have positive effects on the children of low-wage workers? We examine this question using data from the National Longitudinal Survey of Youth.||
The RA will use Stata to construct tables and graphs, will review relevant previous research and will assist with editing of academic papers on these topics. The RA will be supervised by Carl MacPherson, an advanced Ph.D. student in Economics
|12||yes||BPP Haas grad student. Applied for mentoring training in August||2 new||Political Economy of pollution||The study aims to explore how electoral incentives can lead to local politicians choosing dirtier growth oriented policies, prioritizing economic growth at the expense of environmental outcomes. We aim to test relevant hypotheses in the context of India using satellite based data on pollution, night time luminosity, agricultural fires etc.||
RAs would get experience extracting data from official websites and converting them into usable formats. They would also get to learn how this data can be used to establish causality of hypothesized relationships through some analytical exercises in Stata and/or python. Some literature review might also be on the table
|13||yes||BPP (Haas)||2 new||Outsourcing public healthcare delivery||The objective of this research project is to explore whether or not outsourcing of public healthcare delivery in the context of a developing country can lead to better outcomes, and why. We intend to test relevant hypotheses using data on health outcomes in Pakistan where multiple provincial governments experimented with the model of outsourcing delivery of public healthcare to the private sector at the district level in a staggered fashion.||
RAs would be required to find and scrape/download relevant data from the internet, and then work with the PIs to clean it and produce basic summary statistics. RAs would also be expected to join brainstorming sessions with PIs for isolating and addressing holes in hypothesized narratives that emerge as part of the process.
|14||yes||Econ grad student mentor||3 new||Labor Markets and Technology Adoption in Burundi||Farmers in Burundi practice mostly subsistence agriculture, with productivity levels that rank among the lowest in the world (World Bank, 2019), making improvement in this sector a first order priority. While technologies exist that would seemingly improve farmers yields, takeup of such technologies is surprisingly low. In this project, we explore the role that constraints in the labor market might play in slowing technological diffusion. We run an RCT in partnership with a local NGO that aims to test several interventions that will reduce labor market frictions, to explore the impact on technology takeup.||
The job will primarily involve working with data. This includes writing cleaning and analysis code in stata, that will be used to explore incoming survey data. Prior experience with stata is not required, however a willingness to invest a minimum number of hours each week learning good coding practices is required.
|15||yes||Econ grad student mentor||1 new||Are Global Growth Forecasts Systematically Wrong?||Growth forecasts made by banks or international institutions like the IMF are a crucial input into international flows of capital, which have important implications for developing countries. However, professional forecasts may be wrong in systematic ways—for instance, China’s forecasted GDP growth has consistently undershot the reported official numbers. We are exploring the nature of these forecast errors, and if—because of behavioral biases, political motivations, or other factors—there are fundamental reasons why professional forecasts are wrong.||
Assisting with data cleaning in Excel to build a database of global forecasts. Preliminary analysis and chart-making.
|16||yes||Econ grad student mentor||2 new||Unlocking the East Asian Miracle||How did East Asia escape poverty and become one of the world’s most developed regions? We are studying the roots of industrial transformation in Taiwan, South Korea, and now China, with a focus on two theories: first, demand-side shocks from the US; and second, agrarian reforms in the countryside. By harnessing newly digitized archival materials, and bringing in novel sources like satellite imagery, we hope to advance our understanding of the East Asian miracle, and learn lessons to apply to today's developing world.||
Cleaning/coding data. Working with GIS data. Running preliminary analysis in Stata/Python/R.
Not required, but knowledge of deep learning/neural networks a big plus. Similarly, not required, but Chinese reading ability is a plus.
|17||yes||Econ grad student mentor||3 new||Understanding Worker Responses to Advance Notice of Job Loss||How do workers react to advance notice of layoffs occurring at their firm? In the United States, workers are legally required to receive forewarning of large plant closings or mass layoffs, ostensibly so that they can make adequate provision for future job loss. This project will 1) first build a database summarizing such notices across the country; 2) examine firms’ compliance with advance notice policies; and 3) leverage novel administrative data to understand the effect of job loss forewarning on worker outcomes. We hope to incorporate these empirical insights into developing new models of worker mobility and job transitions.||
Mass layoff notices are generally available from state workforce agencies by request, but often need to be cleaned and standardized for research use. In addition, laws governing these notices often differ across states and over time. Under close supervision with us, RAs will be responsible for:
• Developing a companion database of how state-level regulations have changed over time
• Cleaning firm layoff notice records, organizing data into a workable format for analysis
• “Fuzzy” merging the data to other administrative datasets
• Conducting validation tests on the constructed data
This project would entail close collaboration with us, and so we are looking for the following skills:
• Strong analytical skills and programming skills, including fluency in Stata, R, or Python (multiple languages are a plus)
• Ability to work independently
• Background or interest in labor economics, public economics, empirical macro, or household finance a plus
• Prior research experience a plus but not necessary
|18||Econ faculty||2 new||Behavioral Economics and Behavioral Finance research focusing on experience-based learning||How does lived experience affect financial decision making? Experience effects (or, experience-based learning) is a relatively new concept in economics, which postulates that our past lifetime experience shapes our beliefs and risk attitudes for years and decades to come. For example, I’ve shown that if you consistently see increasing prices at the grocery store you will believe in higher inflation levels which will affect your retirement savings level. I’m working on several projects on this topic including ones relating to the pandemic, politicians, and banking risk management. Students may also work on some of my other behavioral papers.||
(1) Performing parsing and data analysis such as regressions (requires experience with programming in one of Python, R, Stata, etc.) (2) Finding new data sets (3) Reviewing related economics, and occasionally psychology, literature (4) Identifying new applications of "experience effects", potentially beyond finance and macro
|19||Econ faculty||4 new||Pandemic Aging||In this project, we aim to analyze how stress induced by Covid-19 affects people's health outcomes, specifically aging. To this end, we make use of machine-learning based apparent aging software, a large dataset of facial images of people, and variation in the severity of the pandemic across space and time. The analysis builds on existing work documenting adverse health outcomes in terms of accelerated aging and reduced life expectancy for stressed CEOs.||
data collection, data cleaning, data analysis, idea brainstorming
|20||yes||Econ grad student mentor||4 new||Turning pictures into data||I work on several projects about the long-term consequences of past events and policies. They include the impact of Stalin's repressive policies and Soviet reforms regarding female labor force participation. The challenge with such research is that the data, if exists, is provided in a form that is impossible to analyze directly (paper document, pdf, photo). Therefore, turning historical data into a spreadsheet format is a necessary step. With the research assistants’ help, I would like to turn several historical documents into a format ready for analysis by applying modern data digitization methods.||
I am looking for students who are 1) eager to learn how to apply modern tools to digitize data (e.g. OCR programs, Python text recognition modules, etc.) and/or 2) interested in cleaning and processing the resulting datasets to prep them for analysis.
Depending on RA interests and skills, they can contribute to activities with different levels of difficulty and learning required -- 1) digitizing documents (by applying machine learning methods or using OCR programs), 2) cleaning data in Excel, 3) cleaning and merging files in Python, Stata, or R.
Ability to understand text in Polish/Ukrainian/Russian would be very useful for some tasks.
Prepared by Martha Olney (email@example.com)
Last update 8/22/2021 5:15 pm