Assar's Ten Commandments

What is the best way of running a research institute? Every researcher probably has his/her own views on this issue, i.e. about the requirements, and probably also definition, of success. My own experiences, after twenty-five years as head of the Institute for International Economic Studies (IIES), may be summarized in ten simple principles.

1. A research institute should try to reach the international research frontier and, indeed, contribute to push that frontier ahead. This is the only reasonable way of defining excellence in research and, hence, to avoid mediocre or even poor research. Indeed, poor research is worse than no research at all, as the former type of activity spreads misconceptions that competent researchers subsequently have to spend time fighting.
2. For such ambitions to be fulfilled, it is important to publish internationally, especially in prestigious academic journals, so that the research is evaluated by the international research community. Otherwise, there is a severe risk that the ambitions of the researchers become constrained by a domestic, and hence lower, level of aspiration. It is very difficult for the leaders of a research institute to hold back publishing of mediocre work if it has its own publication outlet. The IIES has, for this reason, abstained from having its own printed publications.
3. It is important to have outstanding foreign visitors at an institute – to import knowledge and to give a further boost to the international aspiration level. During the 1970s when we were mainly a research-training institute, a number of foreign visitors who stayed for six months or even a year, were extremely important for the gradual upgrading of our competence. Today, when the Institute has seven full professors (as compared to one before 1984), the importance of having visitors staying for long periods of time is somewhat smaller.
4. Every researcher at good institutions tends to develop his/her own international research network. Politicians and university administrators often believe that research networks should be organized, for instance by agreements and contracts between institutions. This is, in my opinion, a completely wrong approach.Different researchers need international contacts in different parts of the world, and formal agreements between universities are usually unhelpful in bringing this about. Such contacts may instead start by a visit from a foreign researcher, or by a visit by the researcher to a foreign institution. Successful international networks are reflected both in intense individual interaction and in joint papers, rather than in formal agreements between institutions. Participation in international research projects and conferences, and the organization of such activities, is another way to
   participate in such networks. To provide financial support for activities of this type is the best way for politicians and university administrators to stimulate international research co-operation.
5. It is important to concentrate the activities to a limited number of areas. During the first ten years, the IIES concentrated its work in one field only, macroeconomic theory and economic policy for open economies. This choice of topic occurred mainly by the power of example, as I and a few students who came to the Institute in the early 1970s happened to be interested in that issue. Others interested in the same field joined the institute, and a coordination of research activities took place as by an invisible hand rather than by conscious planning. The institute has subsequently widened its scope to a few new fields, such as international trade, labor economics, industrial organization, economic development and international finance, as the research staff has expanded.
6. Successful research also requires that a critical mass is reached in one or a few areas. It is my experience that this requires at least half a dozen people in each field. I then define a critical mass, operationally, as a situation when there is a high probability that a few people at the institute are truly interested in reading each other's drafts. Seminars also become much more interesting if at least three or four people are interested in the same field. The frequency of joint work among people at an institute is an indicator of whether a strong interaction has been achieved, and hence whether a critical mass really exists. It is, from that point of
   view, encouraging that much of the output of the Institute today consists of joint papers. It is also useful to have a mix of generations - senior researchers (professors), post-doctors and Ph.D. students.
7. A strong interaction between theoretical and methodological work, on the one hand, and applied research, on the other, is important for successful research. Our institute was earlier often criticized in Sweden for emphasizing theory too much. "Irrelevant work" or "space research" were terms used about us by the chancellor of a Swedish academic institution with a more applied profile. This emphasis on solid theoretical and methodological training has, however, been a key factor behind the contributions of the institute also in applied research. Without a strong theoretical and methodological base, applied research tends to be mediocre or worse, and also to deteriorate over time. There is, however, also a reverse causation: applied researchers give important impulses to the more theoretically oriented ones, and challenge the empirical relevance of their work.
8. Employment decisions are the single most important administrative decisions at research institutes. If the right hirings have been made, a director can concentrate on his main task, that is to create an atmosphere of enthusiasm. (Occasionally, he should, of course, also recommend researchers who have lost steam to move to other places.) To keep administration to a minimum, it is also important to let everybody "administrate himself" and to distribute joint administrative tasks to different members of the institute. High-quality research requires that everybody has the right to choose his/her own field and topics. Command and hierarchies do not fit a creative research environment: everybody should be his/her own boss. It is also very important to have competent secretaries and research assistants, which our institute has always emphasized. Not least the foreign visitors have appreciated these basic services at the institute.
   
9. It is a great advantage for a research institute to be part of a good university. This contributes to new intellectual impulses from other departments and disciplines. It also provides interaction between research and education, which is stimulating both for researchers and for students. The members of the IIES have in recent years increasingly been engaged in teaching at all levels in the Economics Department. We have also tried to convince our colleagues in various departments how important it is that full professors engage themselves in teaching also on the undergraduate and master’s levels.
10. Finally, a research institute may want to produce articles and books not only for the international research community, but also for laymen interested in economic issues. This is, indeed, one way that a tax-financed research institute may give a rather direct, and highly visible, return to tax payers. Members of the IIES have done so, and in this context dealt with problems both in Swedish economic policy and contemporary international economic issues. Experience shows that it is possible to make such contributions without a serious loss of time for academic research, as long as they remain a side activity. In fact, close contact with discussions on economic policy issues help researchers to find suitable topics for academic research.