Syndicate content

Doing Development Economics at a Liberal Arts College Part One

David McKenzie's picture
With the job market coming up, me giving a talk to a great group of faculty and development students at Williams College last week, and seeing a program for the recent LACDEV conference, I thought it might be interesting to learn a bit about life as a development economist at a liberal arts college. I asked four faculty at different schools for some thoughts, thinking I might get two to agree, but was very pleased to get excellent insights from all four. They noted there is considerable heterogeneity among liberal arts colleges, so I am delighted we can share these different viewpoints. I am therefore breaking this up into two posts, with the second post tomorrow.

Today’s guests are Kartini Shastry (left), an associate professor of economics at Wellesley College, and John Maluccio (right), professor of economics at Middlebury Economics (who notes he crowd-sourced also from his other development colleagues at Middlebury).

          

What are some of the differences or possible challenges of doing development research at a liberal arts college compared to a research university?

Kartini Shastry: Liberal arts colleges offer economists a different balance of teaching and research than traditional research universities. My teaching load at Wellesley, 2-2, is not uncommon among research universities but there is more faculty-student interaction (during office hours, for example) and promotion decisions weigh teaching quality more than those at research universities. That said, research productivity is a critical component of promotion decisions. In the economics department at Wellesley, we take into account the greater teaching-related demands on faculty time by trying to emphasize research quality, but accepting that we may write fewer papers than we would at an institution where teaching was weighted less. We strive not to sacrifice quality; my development colleagues have publications at top-tier journals, including the American Economic Review, Review of Economics and Statistics, and Applied Economics Journal: Applied Economics. Another aspect of research requirements at small liberal arts colleges that helps account for the effort spent on teaching is that having a ‘top five’ publication is not usually necessary for tenure. I found that this freed me to focus on doing high quality research that contributes to the literature without the anxiety induced by worrying that I should ditch projects that were not ‘top five’ material. 

A critical input to research productivity is exchanging ideas with other researchers.  Since small liberal arts colleges are, well, “small,” one may not have a large array of colleagues with similar interests on campus, or nearby. One may need to be a little persistent to get invited to give seminars, or be on conference panels. On the flip side, there is a community developing among researchers at small liberal arts colleges, manifesting itself most formally in the annual Liberal Arts Colleges Development Economics Conference (LACDEV). I have found this conference to be a supportive and productive venue to present work, discuss others’ work and meet potential co-authors.

An obvious difference between working at a small liberal arts college and a research university is that we don’t have PhD students. This has some disadvantages, but also some advantages. Not having access to PhD students as research assistants and a source of fresh ideas are probably the biggest disadvantages, but teaching and advising graduate students places non-trivial demands on faculty time at research universities. Personally, a related blessing for me is that I have fantastic senior colleagues at Wellesley who have had time and energy to mentor me and read my papers, which they may not have had if they were also responsible for graduate students.

So far, I’ve discussed differences that are important, but not particularly field-specific. As an applied micro/development economist, evaluating natural experiments using publicly available data or even collecting data in the field, for example, from secondary sources feels no different at a liberal arts college than it would elsewhere. Being at a liberal arts college does make managing field experiments more challenging, however. Teaching both semesters means I’m not free to spend weeks in developing countries during much of the year; and buying out courses frequently is not common. (That said, many small liberal arts colleges have generous junior leave policies.)

John Maluccio: Seems fair to say that keeping up with the literature is more challenging at LACS with typically greater teaching responsibilities. In our department, some faculty form small reading groups for this purpose. Another approach is through your classes where you tend to have a lot of latitude, precisely because you are not training PhDs. While few of teach top journal papers at a technical level, all of us teach seminal papers (eg. in senior level Econ Development survey-type courses), putting the emphasis on the big questions asked and broad strategies used to answer, and relatively less attention paid to how the standard errors were calculated. Varying the papers each year provides an opportunity for you to study important papers more carefully. 

If you’re keen to work with graduate students, advise PhD theses and have a high-level weekly seminar on development, then it is true that most LACs are not your best fit. For some, it is possible to occasionally advise PhDs at other universities, which can be both valuable to the faculty member and is valued by the institution as service to the profession.
It was indeed recognizing some of the challenges of staying connected and soliciting critical feedback, more salient in smaller departments with less potential interaction with other development economists, that led several assistant professors from a set of LACs (see website for list) to begin the Liberal Arts Colleges Development Economics Conference (LACDEV) begun five years ago at Amherst. Middlebury was not part of the initial organizing group but the conference has certainly benefited our department and has led to a number of fruitful collaborations.

Research expectations are heterogeneous across LACs, but one feature that may be appealing to some development economists is almost by default they tend to value research across different fields; one typically need not publish only in mainstream economics journals. For example, during my tenure review it was clear to me that some of the papers valued by my department were those published in public health. As publication outside your specialty is valued (but not required), there is also less of a need to distinguish yourself as the world’s expert in a single topic. I believe this has allowed many of us to broaden our research interests in ways that may not have come as easily in a research institution.

Teaching responsibilities certainly depend on the size of your department, but in larger ones like Middlebury we are able to specialize. Typically, a faculty member will teach something in the core (eg. metrics or micro/macro theory) and then one or two preps in their area of interest. Especially at the upper levels, given the liberal arts mission, there is scope to design courses that will be appealing both to students and to the faculty member and allow you to read/learn/etc in areas of interest to you. These courses can also allow you to refine your own thinking on “big picture” important questions.
Teaching loads differ, at Middlebury it is approximately 2-2 one year and 3-2 the next. That said, with specialization this almost never translates into as many preps, though that can be possible in smaller departments. What is true is class sizes will almost certainly be smaller – and class engagement more intense, possibly in class and probably outside class. Indeed a metric considered in assessment of faculty at Middlebury is “teaching outside the classroom” a broad construct that includes advising or other engagement of students outside the lecture hall (including RA work). Actively, and deliberately, thinking about best practices in teaching is also important.

How have you managed to overcome some of these challenges, and what are some of the additional advantages or ways being at a liberal arts college has helped you as a development economist?

Kartini Shastry: Personally, I have tried to deal with this challenge by working with co-authors at research universities who recruit their graduate students and by relying on full-time research assistants working at places like J-PAL or IPA. These strategies have worked to some degree, in that I have managed a few field experiments; but here the numbers game may be a bigger problem – if a fraction of field experiments get published, working on only a few can be risky. Before tenure, I followed a mixed strategy – managing field experiments, while also exploiting natural experiments.

I would be remiss if I did not note that the undergraduates at Wellesley and other small liberal arts colleges are intelligent, hard-working, and creative. Our top students are just a year or two shy of attending top-tier graduate programs in economics. These students can sometimes substitute for graduate students as research assistants and co-authors. Many of my colleagues and I have published papers co-authored with our students in top journals – for example, Pinar Keskin and I published joint work with Helen Willis ’14 (currently a graduate student at Berkeley) in the Review of Economics and Statistics (see Levine, McKnight, and Heep, American Economic Journal: Economic Policy 2011 and Frydman, Hilt and Zhou, Journal of Political Economy 2015 for more examples of papers joint with Wellesley undergraduates). Going beyond research productivity, teaching smart, engaged undergraduates can be fun and rewarding.

John Maluccio: At Middlebury, there are resources for research that allow a reasonable amount of travel and covering the costs of student RAs, but funding for larger projects must typically be sourced from outside grants. We have a grants office offering valuable support, but it is fair to say that more of the logistics of grant submission are the responsibility of individual faculty members. Notably, some granting organizations give extra points for LACS, in particular the NSF RUI (Research in Undergraduate Institutions) program.

As a highly selective LAC, Middlebury also has an ample supply of bright undergraduates bringing various skills (e.g., language, CS, GIS, basic econometrics but typically not high level experience with data or modeling) eager to gain research skills. In assessments, the college particularly values when working with RAs leads to substantial training or joint publications. 

With the different barriers to organizing large new research projects in mind, a model that I have found useful, in particular in the context of our thesis program, has been to engage colleagues from other institutions involved in new projects and pose the following question. Are you working on a project for which there are perhaps some (high risk) secondary research questions you’d like to explore, but realistically won’t have time to? With guidance, let’s hand the data and question to an undergraduate to play around with for a term and see where it goes. If the project goes nowhere, it remains a valuable thesis project – which is mainly about process and learning. If it seems to have legs to go forward, then I explore with the colleague whether to consider coauthoring an extension. 

Like many other LACs, Midd offers a pre-tenure year-long sabbatical in part recognition of the difficulty of carrying out research during teaching years. Combined with summers, this enables many of us to spend substantial time in the field at particular points.

Any advice for development economists on the job market considering liberal arts colleges as a place to apply to/take up a job offer at?

John Maluccio: Do your research on the department, certainly before ASSA and even better before preparing your cover letter. Understanding some of the main characteristics of the academic setting of a place and some of its people (e.g. for Middlebury the feature that we have a 4-week winter term during which faculty can teach or that we have large joint programs with other departments) can help you think about and frame your response to an inevitable question about what course you’d like to design and teach. This is a question taken seriously. The type of comment that would not be viewed favorably in a cover letter, for example, is something like “I look forward to working with graduate students.”

Think about your job market presentation as an opportunity to showcase your potential teaching skills, i.e., demonstrate you would be a good teacher. Typically, the audience will include undergraduate students as well as non-economist administrators. This doesn’t mean pitching only to them (you could lose the interest of the department) but thinking about how to appeal to a general audience is valuable.

Kartini Shastry: If you are interested in a job at a small liberal arts college, you need to signal that you’re really interested. We are wary of hiring someone if we don’t think the candidate is going to be happy and stay. So you need to signal that you would be happy with that school’s specific balance of teaching and research. If you’re considering whether to take a job offer, you’ll want to think about whether you actually would be happy with that balance. Personally, I really enjoy teaching the very talented, engaged young women at Wellesley and I appreciate the fact that the effort I put into both teaching and research is seen and valued.  
 
Look out for part two’s post tomorrow, where we hear from Jessica Hoel and Tahir Andrabi
 

Comments

Submitted by Jennifer Fievre on

I went to Williams and took a couple of classes in Econ. For one of them, we had to visit the CDE. I remember a very diverse class who loved the program. They complained about 2 things: Williamstown being a bit isolated and the cold weather. Otherwise, the CDE students were very happy with the program.

Add new comment