A Guide to Choosing a PhD Topic (Part I)

There are a handful of excellent guides to different aspects of the academic profession that are invaluable to PhD students in economics. See here for a comprehensive summary.

As a complement to these guides, I want to focus on one specific (and often under-appreciated) choice that a young researcher makes: the choice of the PhD topic. Most students considering embarking on a PhD focus on more-immediate choices such as the institution and the choice of PhD supervisor rather than the choice of topic. All three choices, however, have persistent long-term effects on your career – and there is no good reason to under-estimate the importance of the choice of PhD topic.

In contrast to the other guides, I provide more concrete guidance on what constitutes a good research topic and spell out why the choice of research topic matters. Specifically, the choice of topic will: (partly) determine how quickly you are able to complete your PhD; influence the type and quality of job opportunities you face after your PhD is completed; shape your future research trajectory; and be an important component of your academic reputation. For those interested in pursuing a research career in academia after your PhD, you should care about all of these factors. For those interested in other career options – government, international agencies, consulting firms, or central banks – at least the first two will be of interest.

In my view, people systematically under-value the skill associated with picking good research topics. Perhaps part of the reason for this is that people don’t recognise that the choice of PhD topic isn’t really a one-off – although most people only choose their PhD topic once, the lessons learnt here have lots of useful future applications. For academics, this skill is applied to the ongoing set of research topic choices made throughout your career. And for the non-academic economist, there are similar long-term benefits associated with being able to clearly frame (and motivate) the project you are working on. In my experience, this skill is scarce and valuable.

Note that there are two levels to the discussion: the ‘research topic’ (e.g. government outsourcing) and the more specific ‘research question’ (e.g. does outsourcing save money?). When I refer to the ‘PhD research topic’, I literally mean the set of research questions that underpin the PhD dissertation. For academics, the ability to select good research questions has important consequences for the likelihood of getting published and the impact of your work once published. Fortunately, it is something that you can improve over time with practice.

PhDs come in all shapes and sizes: some are book-like while others are essays-on-a-related-theme; some are pure theory while others are purely empirical. There are many factors that will affect the likelihood (and speed) of PhD completion: student-specific factors (e.g. intelligence, perseverance), university-specific factors (e.g. support for PhD students, size and quality of PhD student cohort), supervisor-specific factors (e.g. ability, commitment to students, interest in your topic), and the choice of research topic (e.g. novelty, importance, tractability). I assume away the first three factors and focus purely on the choice of PhD topic.

A useful framework to apply to the choice of PhD topic is ‘search theory’: students trawl through the set of available potential topics (with incomplete information) until they find one that satisfies their preferences. Delaying the choice is costly since the longer the time taken, the further behind the student lags. But making a hasty decision can end up having painful consequences in the long term. So, there is an optimal point at which point the student trades off the costs and benefits and decides to start on a particular research topic. Given this, it is important to keep asking yourself the question: is this topic good enough (relative to the other alternatives) to stop the search process?

The first reason why choice of PhD topic matters is that PhD students asking ‘good’ research questions will finish faster than others, all other things being equal. Many PhD students start off with grandiose questions that are impossible to answer in a 3-year period. A PhD is not about answering the world’s big questions; it is about demonstrating that you have the necessary skills to do independent research. This includes demonstrating that you have command over a wide range of research techniques and a deep understanding of all aspects of the research production process. Once you realise this, it will be clear that asking a ‘good’ research question enables you to showcase these skills: to articulate the problem clearly, to explain why it is interesting/important, to utilise appropriate tools and techniques to answer the question, and to draw sound conclusions.

There is also good reason to argue that your PhD topic will influence your future job opportunities, research trajectory and reputation. Once you finish your PhD and enter the job market, a complex matching process occurs: universities will be looking to strengthen in certain areas of the discipline where they are weak (or where they see a latent demand), and PhD students will be looking to rank their preferred institutions based on their location, quality, teaching load, and salary. Where you end up getting a job will be the result of this matching process.

NOTE: this is an abridged version of a paper forthcoming in the December 2013 issue Australian Economic Review. See here for the complete version (paywalled).