Description

Infrastructure bottlenecks in transport, energy, and telecommunications, as well as the social layers of health, education, water, and agriculture, remain the primary hurdles for sustainable economic growth in developing countries. These systems are not merely physical assets; they are the fundamental conduits for expanding productivity, alleviating poverty, and fostering human capital formation. In the modern era, the definition of infrastructure has expanded to include Digital Public Infrastructure (DPI) such as biometric identity systems and digital payment stacks which have the potential to revolutionize public service delivery, reduce price dispersion, and integrate marginalized populations into the formal economy.

Beyond digital realms, cities and rural areas continue to evolve. From rural electrification driving structural transformation to urban transit systems reshaping labor market access, infrastructure investments are the primary levers for economic integration. Understanding these complex changes requires a multi-faceted approach that bridges different sub-fields in development economics.

This tutorial provides the toolkit to navigate this landscape using frontier empirical research. We will move beyond passive reading to engage in critical synthesis, utilizing modern empirical research methods to analyze data and answer vital questions. Our objective is to develop the technical skill set and theoretical intuition necessary to design, execute, and communicate original research in a world where data is abundant but causal clarity is paramount.

Goals

• Critical Synthesis: Learn to read and critically evaluate frontier research
• Empirical Toolkit: Master the application of causal identification strategies to evaluate impacts of infrastructure
• Data Proficiency: Develop the ability to clean, analyze, and interpret complex data sets
• Original Inquiry: Identify a unique research question, navigate the “data revolution” to find or scrape appropriate sources, and execute an original research project.
• Scientific Communication: Enhance the ability to communicate technical economic findings clearly, maintaining rigorous standards for academic honesty

Prerequisites

• Microeconomics: Students should be comfortable with constrained optimization and market equilibrium concepts. All Ec 970 tutorials require prior successful completion of Ec 1010a/1011a.
• Statistics: Beyond basic probability, students should have an interest in understanding concepts related to causal inference. All Ec 970 tutorials require prior successful completion of one of the stats courses approved for the economics concentration.

Evaluation

Research Writeup, 4 pages per module of the course (15%)

An empirical economics research paper typically has the following sections in some form:

• Abstract and Introduction
• Context and Background
• Literature Review
• Research Design
• Data and Results
• Conclusion

For every module of the course except Communications, you will be provided with a prompt to write one of these sections based on ideas related to the papers you would have read as part of the previous module. The idea is to give you some practice and feedback with different parts of the writing process.

Midterm Exam (20%)

Thursday, October 22, in class.

Throughout the semester, we will read and discuss many papers. In the midterm, I will provide excerpts from one of the papers we have already read, in the form of a “reading packet,” and will ask you to answer several short-answer and short-essay questions regarding the paper - I could ask what the research question is, the paper’s place in the literature, questions on the theoretical and/or empirical approaches, data, results, and/or weaknesses.

Little active preparation for the midterm is required. I emphatically do not want you to memorize all the notes you have taken in class. Instead, I expect that by doing all the readings and being fully present and participatory in class, you will organically pick up the skills required to do well in the midterm. That is, you will practice and understand how to read and extract information from journal articles, and how to question them. These are the skills you should bring to bear in the midterm. I would not, for example, ask what the findings of a paper are without providing you with the necessary tables, in the reading packet, to answer that very question. I would not ask you what other papers have written on the same question, without providing in the reading packet a brief literature review. I would however expect you to be able to look at the empirical methods laid out in the reading packet, and state what potential weaknesses the methods have. In other words, no outside information (no memorization!) should be required to successfully complete the midterm, but your critical thinking skills will be required.

Students with DAO accommodation will use the Testing Center to complete the midterm.

Paper Presentation (10%)

Presenting your own or synthesizing others’ work will help you not just understand the material well but also help you think about how to communicate it clearly and concisely to an audience. To that end, I want each one of you to lead the discussion on one paper in the semester.

Hence, each student will present one paper on the syllabus to the rest of the tutorial group. This includes summarizing the paper’s empirical strategy and findings, anything you particularly liked/disliked/found confusing, and discussing any interesting insights your peers had in class. This also means that you try your best to answer any questions that your peers might have in relation to the paper that you chose to present.

The presentation schedule is on the Canvas syllabus page.

Participation and Intellectual Engagement (10%)

Attendance and active engagement are the cornerstone of this tutorial. In an era of automated summaries, the ability to verbally synthesize complex economic arguments and debate them in real-time is a vital professional skill for any economist. Participation is not just about speaking; it is about listening and responding to your peers.

Expectations for Engagement

• Presence & Preparation: Attendance is mandatory. You are allowed two “no-questions-asked” absences; beyond this, your grade will be impacted unless the absence is certified by a doctor or resident dean.
• Active Contribution: You are expected to arrive having read all required materials and prepared to contribute to every session. This includes asking both clarificatory questions (to ensure technical understanding) and substantive questions (to challenge empirical designs or theoretical assumptions). To ensure genuine engagement, I will frequently ask for your perspective on a reading we discuss in the class.

Stata/R Problem Sets (10%)

A necessary skill you should learn during this tutorial is how to work with data, including building or cleaning datasets as well as analyzing them to answer questions that you are interested in. This involves mastering the skill of statistical programming. Economists commonly use Stata or R for their code. You can choose which language you want to learn and/or use. If you wish to speak with me about statistical programming, I am more comfortable with Stata than R. However, you are still free to learn Stata or R in the statistical software minicourse.

Like all Ec 970 tutorials, this tutorial’s assignments include four problem sets from the mandatory statistical software minicourse. It is important that you carefully attempt these as they are essential for your final paper. You will need to complete and turn in the code, log files and output.

Research Paper and Presentation (30%)

Besides learning how to read and reflect on papers, this tutorial aims to help you write papers. Every aspect of the tutorial, from reading papers, discussing them with peers, working on problem sets – all are meant to help you reach a point where you can find a way to address a research question that is important and/or that interests you.

Therefore, the main assignment is writing a term paper on any topic of your choosing. The assignment will be broken into multiple smaller assignments, as detailed below. Note that it is more important that you write a good paper addressing an interesting topic than it is to write a paper on infrastructure related topics. Overall, a good paper would likely cover one of the below, but I am open to alternative suggestions.

1. An empirical paper using existing data. The section on data below highlights several data sources you could use. The paper does not necessarily have to be on infrastructure if it’s a question that interests you and helps you write a good paper. The paper should attempt to answer a causal question using empirical strategies like most papers we cover in this tutorial. If you do not have a causal identification strategy, you should clearly discuss limitations to a causal interpretation of your results.
2. An empirical paper using original data. You could create your own dataset, for example by web scraping the data you need.
3. Replication and extension paper. You could replicate a paper using data posted by authors (not always available or complete, so plan to search and attempt early on). In addition to replicating, you should plan to extend the paper, for example by using a more recent technique, looking at additional outcomes (clarifying motivation/theory behind why the extension might be interesting), etc. You are most likely to find papers’ data, if it exists, on the authors’ website or the publishing journals’ website, and sometimes on Harvard dataverse, ICPSR, or IPA/JPAL/AEA dataverse, all of which have been linked below.

Any of the above choices for a research project will require a lot of work and time, so you should start early on and should discuss with me your plans for additional resources and guidance.

The assignment will be broken down to pieces to help you keep on track.

1. Early-stage Research Ideas (3%, at least two paragraphs per idea): Please submit at least two early-stage research ideas for your final papers (in one submission). Each early-stage research idea must have a brief introduction, research questions and potential data sources. It may include empirical strategies/methodology and an early look at the related literature. Do not worry about having a very polished idea and proposal at this stage. After you select one research idea to move forward with, you will develop it further into a research proposal. Note that finding a hypothesis testable with available data can be challenging, so start thinking about your research topic and the available dataset early.

2. Proposal (3%, 5-6 pages): Your research proposal should elaborate more on your selected early-stage research idea and provide a roadmap for your final paper. In choosing which idea to move forward with, it is crucial to consider the data availability and feasibility given the time constraint. You may also choose an entirely new idea that is different from any of your early-stage ideas. Your proposal should contain the following components, although the order and structure can be flexible, and you may combine several components into a section:

   1. Motivation/Introduction: Start with a brief introduction for the paper and why the topic is important. You can refer to the first paragraph of an economics paper to see examples of this.
   2. Research questions and hypothesis: Formulate clear and specific research questions your paper seeks to answer and discuss your hypothesis. Since this is an empirical research assignment in economics, for this class, be explicit about whether you are interested in answering a causal relationship question or a correlational relationship only.
   3. Context: If necessary, explain the context of your study. For example, if you are evaluating the effect of a specific program or policy, describe the policy, including the details that would be relevant to your study. You can combine this section with the motivation/introduction or data section as you see fit.
   4. Methodology: Describe the empirical strategy you will use to answer your research questions. Be as specific as you can. For example, if you want to use difference-in-difference (DID), be specific about who or what you are comparing. If you are interested in using instrumental variable (IV), be specific about what your instrument is, what your assumptions are, etc. If applicable, it may help to have a regression specification at this stage.
   5. Data: Outline the dataset you plan to use, including the indicators that you will use to measure your dependent and independent variables. Be explicit about the level of analysis you are planning to do (e.g., individuals, households, zip code, county, country, or others). At this stage, make sure that you can access the dataset in time to conduct the analysis for your paper.
   6. Literature: Provide a brief overview of the existing literature related to your topic. Briefly discuss how your research draws from and contributes to the literature.

3. Revised Research Proposal (4%, 3-6 pages): This is an opportunity for you to submit a revised research proposal that incorporates and addresses the feedback from your prior submission. Ideally, this version should be a ready-to-execute roadmap for your final paper.

4. Research Paper (15%, 15-20 pages, not including tables, figures, and references): Your research paper should include the updated and revised contents of your proposal – after incorporating the feedback you receive – and your complete analysis, findings, and discussions. We will post a longer guideline on the final paper on Canvas. Please submit your final paper on Canvas and include the codes used to clean the data and perform the analysis in a replicable format. Unless you use proprietary data with a limited data use agreement, please also include the dataset – you can email this to me if it’s too large to upload on Canvas.

5. The presentation (5%): This is an opportunity to practice communicating your research, receive comments and feedback from your classmates, and provide feedback on your classmates’ research (the latter will contribute to your participation grade). In a presentation session during reading period (date TBA), you will present your research in 12 minutes with 3-minute discussions. You will also send written feedback via email to one of your classmates presenting that day. We will determine who you will send feedback to via a lottery at the beginning of the class. Please send the feedback to your classmate via email (cc-ing me) by the end of the presentation day.

To assist you in developing your writing skills, you can use this guide “Writing Economics”. Please refer to this regularly as you work on the assignments and the components of the final paper. It will help guide you when you need any assistance.

Of course, you can always reach out to me if you are facing difficulties.

Oral defense (5%)

During/near the final exam slot assigned to us by the Registrar’s office, you will meet one-on-one with me for a discussion of your research paper. This is not a presentation, but rather a Q&A session. I may ask questions about your research question, your motivation, how your research relates to existing papers, your empirical approach, dataset, the weaknesses of your data and methods, your results, questions that this research raised for you, and how this project relates to the course for you.

Due: sign-ups to take place when the Registrar releases final exam dates and times.

Note on assignments

All written assignments will be posted on Canvas with their preliminary guidelines and due dates. Note that all assignment pages should be double-spaced, normal font (ex: Times new roman 12pt), standard margins. All assignments should be submitted on canvas as a pdf document, email submissions will not be accepted.

Grading

• A [95-100]: excellent work, very good writing (clear, succinct, structured), demonstrate high effort and mastery of the subject, presents relatively interesting ideas or creative approaches
• A- [90-94]: very good work and effort, responds to prompt adequately, makes little mistakes, demonstrates mastery of the subject
• B+ [85-89]: good work, effort, and comprehension; full engagement with the assignment’s requirements
• B [80-84]: fair work and effort; submission might have typos
• B- [75-79]: little work and effort; submission might have conceptual errors, or is hard to follow; work doesn’t follow prompt
• C+, C, C- [<75]: unsatisfactory work, indicates minimal command of the course materials

Policies

Office Hours: Please sign up for an office hours slot by the third week of the semester so I can get to know each of you and your interests better. I’d also encourage you to sign up for office hours to discuss your research ideas and feedback throughout the semester.

Attendance and participation: You are allowed 2 no-questions-asked absences before your grade is affected. All students are expected to participate in the discussion in every session.

Laptop and Tablet Policy: During class, I request you not to use any devices, except to refer to your notes or readings. Please note that any other use of an electronic device may adversely affect your grade after one warning.

Expectations for readings: Most of the readings assigned in this tutorial were published in top economics journals. You are not expected to be able to reproduce mathematical arguments on your own. When reading, focus on the introduction, the empirical strategy and the intuition behind the economic arguments. Try to figure out where the paper fits into the broader economics literature.

Communication: Please email me any questions, feedback, and comments that you have about the class and include [ECON 970] in your subject. I will try to get back to your emails within 24 hours; this time might be longer during the holidays. Please keep that in mind especially as assignment deadlines approach.

Submission: Please submit all your written works in PDF to avoid formatting changes or errors.

Late assignment policy: Assignments will lose 5 points per 24 hours of delay unless discussed in advance. Please discuss extensions with me in advance. Be aware that you cannot get an extension on the final paper, so plan accordingly to ensure you can complete all of it. Exceptions are given in case of a medical or personal emergency if certified by a doctor or a resident dean.

Working with others: You are free (and encouraged) to discuss problem sets, the assigned papers and the final paper with one another. However, you must hand in your own (unique) code and written work in all cases.

AI Policy: Generative artificial intelligence (AI) has many benefits and costs. You are encouraged to research these benefits and costs and to think carefully about what they imply for how you can work effectively and productively in your future life and career. For your time at Harvard, in general, it is worth remembering that you likely decided to invest in your education to develop your skills and abilities to think critically and creatively; relying on generative AI to complete class assignments may rob you of these developments and undermine your investment.

At this specific moment in history, when generative AI technologies are undergoing rapid development and tools are constantly changing, it is hard to codify how best to use generative AI to help you learn. However, two facts are clear: First, it is a violation of the Harvard College Academic Integrity to incorporate into your coursework text produced predominantly by generative AI (similar to copying from a book or article). For example, simply rewriting a draft generated by AI is not acceptable. Second, generative AI can produce false or misleading information. You are ultimately responsible for the accuracy of any work you submit. These two facts should inform your use of generative AI throughout this course. Please also be aware of the University’s initial guidelines for the use of generative AI tools, available at the link here.

If you use any generative AI on any written assignment, you must submit a statement detailing how you used it, and what you learned from it. This statement should include all of the ways you engaged AI (i.e. listing all prompts), tracking your workflow sequentially from beginning to end. Note, however, that the use of AI for problem set completion in the Stata/R mini course is strictly prohibited. Violations of this policy will be considered academic misconduct.

Support

Your wellbeing is of utmost importance to me. As a student, you may experience a range of challenges that can interfere with learning, such as increased anxiety, strained relationships, feeling down, difficulty concentrating and/or lack of motivation. I recognize that this environment is challenging and that courses are not the only demanding part of your life.

You never owe me personal information about your (mental or physical) health, or anything else. However, feel free to talk to me about things that you are going through. I promise to listen and, if this relates to your learning, I will work with you to make sure that you succeed in this tutorial.

There are also several resources available to you on campus to support your wellness, including Harvard University Health Services (HUHS), Counseling and Mental Health Services (CAMHS), and the Academic Resource Center (ARC), which offers tutoring services, workshops, and coaching, among other services. If you or anyone you know experiences any high levels of stress, difficult life events, or anxiety or depression, I strongly encourage you to seek support.

Statement on Academic Honesty

Discussing ideas and work-in-progress with others is an important and desirable part of the research process, but in the end, a student’s assignment must be his or her own effort, written by the student, and ultimately based on his or her own thinking. All written assignments must use appropriate citation practices. If you plagiarize, you will fail the course. For questions about Harvard’s stance on academic honesty, consult the Academic Information section in the Harvard College Handbook for Students.

Tutorial Schedule (with major assignment deadlines)

Readings

All readings listed below are required, and each is assigned to a specific session.

1. Education Infrastructure

   1. Sep 15: Muralidharan, K., Singh, A. and Ganimian, A.J., 2019. Disrupting education? Experimental evidence on technology-aided instruction in India. American Economic Review, 109(4), pp. 1426–1460.
   2. Sep 22: Muralidharan, K. and Singh, A., 2023. Improving public sector management at scale? Experimental evidence on school governance in India. Working paper, August 2023. (Earlier version circulated as NBER Working Paper No. 28129.)
   3. Sep 24: Andrabi, T., Bau, N., Das, J., Karachiwalla, N. and Khwaja, A.I., 2024. Crowding in private quality: The equilibrium effects of public spending in education. The Quarterly Journal of Economics, 139(4), pp. 2525–2577.

2. Health Infrastructure

   1. Sep 29: Bennett, D. and Yin, W., 2019. The market for high-quality medicine: Retail chain entry and drug quality in India. Review of Economics and Statistics, 101(1), pp. 76–90.
   2. Oct 1: Kremer, M., Leino, J., Miguel, E. and Zwane, A.P., 2011. Spring cleaning: Rural water impacts, valuation, and institutions. Quarterly Journal of Economics, 126(1), pp. 145–205.
   3. Oct 6: Motohashi, K., 2024. Unintended consequences of sanitation investment: Negative externalities on water quality and health in India. SSRN 4738911.

3. Transportation Infrastructure

   1. Oct 8: Asher, S. and Novosad, P., 2020. Rural roads and local economic development. American Economic Review, 110(3), pp. 797–823.
   2. Oct 13: Faber, B., 2014. Trade integration, market size, and industrialization: Evidence from China’s National Trunk Highway System. Review of Economic Studies, 81(3), pp. 1046–1070.
   3. Oct 15: Banerjee, A., Duflo, E. and Qian, N., 2020. On the road: Access to transportation infrastructure and economic growth in China. Journal of Development Economics, 145.

4. Urban Infrastructure

   1. Oct 20: Tsivanidis, N., 2023. Evaluating the impact of urban transit infrastructure: Evidence from Bogotá’s TransMilenio. Working paper, University of California, Berkeley, August 2023.
   2. Oct 27: Borker, G., 2021. Safety first: Perceived risk of street harassment and educational choices of women. World Bank Policy Research Working Paper No. 9731.
   3. Oct 29: Harari, M. and Wong, M., 2021. Slum upgrading and long-run urban development: Evidence from Indonesia. Working paper, November 2021. (First version July 2017.)

5. Infrastructure, Energy and Climate Change

   1. Nov 3: Gandhi, S., Kahn, M., Kochhar, R., Lall, S. and Tandel, V., 2022. Adapting to flood risk: Evidence from a panel of global cities. NBER Working Paper No. 30137.
   2. Nov 5: Gendron-Carrier, N., Gonzalez-Navarro, M., Polloni, S. and Turner, M.A., 2022. Subways and urban air pollution. American Economic Journal: Applied Economics, 14(1), pp. 164–196.
   3. Nov 10: Chen, S. and Gong, B., 2021. Response and adaptation of agriculture to climate change: Evidence from China. Journal of Development Economics, 148, 102557.

6. Digital Infrastructure

   1. Nov 12: Muralidharan, K., Niehaus, P. and Sukhtankar, S., 2023. Identity verification standards in welfare programs: Experimental evidence from India. Review of Economics and Statistics, pp. 1–46.
   2. Nov 17: Aiken, E., Bedoya, G., Blumenstock, J.E. and Coville, A., 2023. Program targeting with machine learning and mobile phone data: Evidence from an anti-poverty intervention in Afghanistan. Journal of Development Economics, 161, 103016.
   3. Nov 19: Dzansi, J., Jensen, A., Lagakos, D. and Telli, H., 2024. Technology and tax capacity: Evidence from local governments in Ghana. Working paper, June 2024. (Earlier version circulated as NBER Working Paper No. 29923.)

7. Communications Infrastructure

   1. Nov 24: Hjort, J. and Poulsen, J., 2019. The arrival of fast internet and employment in Africa. American Economic Review, 109(3), pp. 1032–1079.
   2. Dec 1: Goyal, A., 2010. Information, direct access to farmers, and rural market performance in central India. American Economic Journal: Applied Economics, 2(3), pp. 22–45.
   3. Dec 3: Blumenstock, J.E., Eagle, N. and Fafchamps, M., 2016. Airtime transfers and mobile communications: Evidence in the aftermath of natural disasters. Journal of Development Economics, 120, pp. 157–181.