HOW TO WRITE A RESEARCH PAPER

1. General reading in area of interest. You first want to get an idea of (i) what have been the main issues in the area that you are interested in, and (ii) how have other researchers answered these questions. To this effect, you should do some general reading (and hopefully you have done this over the summer!). You can find some sources of readings by clicking here. Google searches are another obvious way of getting started. But you may want to check with me on the quality of the material that you find online just to make sure that these are academically solid readings.

2. Define the question that you want to ask. Once you have become familiar with the main issues in your area of interest, you must come up with a question that you want to ask. The question should be specific, relevant, and "answerable" with standard economic methodologies:

Specific/concrete: For example, "The Chinese economy" is not a specific issue or question but "Should the Chinese authorities let the renminbi (the Chinese currency) float?" or "Why have real wages been stagnant in the United States economy?" are specific questions.
Relevant: The question should be relevant for public policy and/or our understanding of economic phenomena.
"Answerable" with some methodology: For instance, "What will happen to the U.S. economy if Democrats lose the mid-term elections?" may be an interesting question for a Time magazine article or for discussion over the dinner table but is not answerable with standard economic methodologies (i.e., we obviously have no data but neither do we have good models to predict events of this nature). The question "Why have real wages been stagnant in the United States economy? or "Why do some financial crises spread from country to country?" are, in principle, questions that can be answered using standard economic methodologies.

You should always ask yourself: What is the question that I am trying to ask? If you have no good answer to that question, you do not have a question!

3. Positive versus normative analysis

Depending on the type of question that you are asking, you analysis can be thought of as positive or normative. Positive analysis means that you are just trying to understand how the world works without taking a stand on how the world should work. For instance, the question "What factors drive the business cycle in the United States?" will lead to positive analysis. You are trying to understand what are the factors that cause business cycles (i.e., GDP going up and down over time) in the United States.

On the other hand, asking the question "Should the Federal Reserve (i.e., the U.S. Central Bank) raise interest rates when inflation goes up?" is a normative question. You are trying to understand how monetary policy should be conducted. (Raising interest rates will probably reduce inflation but at the cost of a recession, so there is a trade-off to be considered.) Of course, it may well be the case that normative analysis can lead to some public policy prescription of a normative nature. For instance, if you were asking the question of what causes business cycles in the United States and concluded that monetary policy itself has been the source of business fluctuations, you would conclude that the Federal Reserve should take a closer look at how it conducts monetary policy.

4. Empirical versus theoretical analysis

Depending on the methodology that you use, your paper can be empirical or theoretical. An empirical paper uses econometric methods to answer the question at hand. For instance, if you ask the question "What were the output costs of the transition from a planned to a market economy in Eastern Europe?" you would run some sort of regression with changes in output as the dependent variable and a bunch of independent variables: control variables (external factors, changes in labor force, changes in capital stock, among others) and then some measure of the change in market institutions that would give you the answer that you are seeking.

On the other hand, a theoretical paper will use a theoretical model to answer a particular question. Typically, you would identify some puzzle that you want to explain with a theoretical model. For instance, suppose that there is evidence that fiscal policy in developing countries has been procyclical (meaning that fiscal policy is expansionary in good times and contractionary in bad times), which is the opposite that you would expect. You might develop a theoretical model that shows that if political interest groups pressure policymakers into spending money when the government runs budget surpluses, then fiscal policy will indeed be procyclical. Alternatively, you could write a purely theoretical paper (i.e., a paper which is not trying to offer a theoretical explanation for some observed phenomenon). For instance, to answer the question “How should the government auction available gates in an airport?” you might develop a theoretical model of auctions that would show that one type of auction would lead to a more efficient outcome than some other type of auction.

Most often, undergraduate research involves empirical analysis, simply because top undergraduates have normally acquired the required basic econometric skills to carry out such a task. In contrast, the typical undergraduate does not yet have the needed theoretical tools to undertake serious theoretical analysis. But if you feel you do and have an interesting question that can be answered with a theoretical model, you should definitely go for it!

5. Write an outline of the paper

Writing (and doing it early rather than later) is a critical component of any research project. One obvious reason is that you eventually need to put into writing your question, motivation, and findings. If your writing is not very good, I would strongly suggest (and we can talk about this) finding some course on writing (believe me, it will be a great investment!). The other (less obvious) reason if that writing forces you to put your ideas in a clear and concise way. Being able to formulate your question in writing will indeed tell you if have a question.

I thus advocate and strongly suggest that you get used to drafting as often as you can. You will first write a one-page proposal (outlining the question and its relevance) and then a 3-page proposal or outline (which will also include a discussion of the methodology to be used).

An outline (i.e, a 3-page proposal) should consist of three main parts:

Introduction: motivates and defines the problem to be dealt with, and says why it is an interesting and non-trivial problem.
Methodology: How will you go about in answering the question you want to address? What economic tools will you bring to bear on the issue? What data will you be looking for?
Conclusions and/or policy recommendations: What do you expect to conclude? How will your conclusions follow from your analysis? What are the policy recommendations that may follow from your analysis?

6. Write a working draft

After you have done the core of the work, you should begin to draft. Don't wait until you have done everything to begin to write the paper. Drafting helps in refining your thinking because it forces you to spell out your ideas and results. It also helps a lot in obtaining feedback from the instructor. At the end of the Fall semester, you should have a roughly 15-page working draft.

7. Structure of the paper

There is not a unique way of structuring a paper but a typical structure would be as follows:

Introduction: here you motivate the question you are asking. This section should contain some charts and tables to illustrate the problem. It should also give an idea of what is the conventional wisdom on the issue and how your paper differs. It should also contain a brief summary of your main conclusions.
One or two sections in which you develop your theoretical and/or empirical analysis.
A final section with conclusions and policy lessons.

The paper should be divided into sections, which should be numbered and titled. Every section must have a point, ask yourself: what I am trying to say in this section?

8. Revise and complete the paper

Having obtained the instructor's feedback, you may need to do some further work and/or refine some aspects of your analysis. You should also think hard about how to improve the overall quality of the paper. With that, you should be ready to revise the working draft and finish the paper. Check carefully the final draft. Work on the style. An interesting, but poorly-written, paper loses much of its appeal. (Revising and completing the paper will be your main task in Econ 397.)

Back to home page