Economic and Game Theory
A Guide to Writing Your First Professional Paper in Economics

by David K. Levine, Department of Economics, UCLA

One of the hardest things an economics graduate student will ever do is write his/her first professional paper. There are a number of common problems that I have encountered over the years, and some generic advice about how to go about writing your first paper.

• It is common not to want to invest a lot of effort in writing a paper unless you know that the results are interesting and exciting. However, it is hard for anyone else to tell if the results are interesting and exciting before you have written the paper clearly. Do not be disappointed if when the paper is finally written it turns out not to be as interesting and exciting as you hoped, and be prepared to invest the time and effort in writing before you are completely satisfied about the substance. Even if the paper winds up not being a great paper, you will be well prepared for writing your second paper.
• The key thing to understand before your write your paper is what the paper is about. This sounds trivial but it is not so. Often papers grow; you have an idea about the way some part of the economy works; or about an interesting extension about an existing line of research. You work out some results. But before you go anywhere, you need to ask - why did I do this and what is it about? What results and insights might be useful to other people? The introduction of the draft of your paper can be short - all the references to the literature that people load these down with can go later. What you do want is a clear statement of what your question is and what your answer is - what the paper is about.
• There is a formula for writing economics papers: introduction, the model, analysis of the model, conclusion. Follow the formula - it is widely used for a reason. Papers often grow from a set of notes you wrote as you worked out your results. These are bound to be incomprehensible to anyone else besides yourself.
  
• Section 2: The Model. Lay out the model as simply and clearly as possible with minimal discussion. Make sure all the notation is defined before it is used; avoid the lengthy equation followed by definition of the symbols in it - no one can understand the equation before you define the symbols. Don't assume because someone else somwhere else used some notation or an assumption that your reader is familiar with it. If things are horribly complicated to explain, most likely the model is too complicated - rethink the model. Can you answer the same question and provide the same insight with a simpler model? Have you constructed the model to be "realistic" without thinking about whether the added details play some significant role in the analysis?
• Section 3: Analysis of the Model. Avoid stream of conciousness - I thought this, then this, then that, then reached this conclusion. Start the section with a verbal description of your question and your answer, plus if you can some intuition about why your answer is the right one. State your theorem first, then do the proof - do not subject the reader to pages of algebra or preliminary lemmas before stating a result. Without some knowledge of the destination, the reader is able only to verify that you are able to do algebra correctly - not of great use to either of you. Make sure you periodically remind the reader what the goal is. Redundancy, repetition and transitional material are good.
  

The best advice of all - the advice given to me by my advisor when I first inundated him with half-written stream of conciousness bits and pieces of algebra - pick up a good journal, Econometrica, American Economics Review, and read a couple of papers carefully to see how they are written.