SciELO - Scientific Electronic Library Online

vol.20 número2El bote de basura como modelo de elección organizacional índice de autoresíndice de assuntospesquisa de artigos
Home Pagelista alfabética de periódicos  

Serviços Personalizados




Links relacionados

  • Não possue artigos similaresSimilares em SciELO


Gestión y política pública

versão impressa ISSN 1405-1079

Gest. polít. pública vol.20 no.2 México Jan. 2011


Sección especial: debates contemporáneos y clásicos


Institutions and Organizations: Differences and Linkages from Organization Theory


James March*, Erhard Friedberg** y David Arellano***


* Professor Emeritus at Stanford University.

** Austrian by birth and education.

*** Professor and researcher in Centro de Investigación y Docencia Económicas (CIDE).


DAVID ARELLANO: Following this wave of new public management that has been so influential in Mexico and Latin America there is a superficial discussion regarding the difference between institutions and organizations. In particular, under the idea that institutions are basically the rules of the game and that, if you design correctly the right incentives, people will follow them and we can change their behavior through those incentives. Therefore, you can build up, you can design and you can create institutions in a rational manner, through these incentive structures. In sum, do you consider that institutions and organizations are different? That you can rationally design them? That by creating the correct incentives you can change the behavior of people in public organizations?

JAMES MARCH: I think that is the kind of question that Erhard should respond to first because he is the expert on local orders and how the same kind of theory works for public and private and multiple networks of various sizes.

ERHARD FRIEDBERG: My impression is that I do not think we are going to have a big controversy around this. My take on this is that this is the language of economists, and that very often this language is over simplifying complicated processes that over estimate the clarity of what is going on. It forgets that you do not command an active system as if it were a machine and you pushed a button and it aligns. So, incentive structures are of course one of the few instruments a policy maker has; so he/she may use them as an instrument but he/she should also be aware of the fact that incentive structures are nothing but incentive structures, and actual behavior will not necessarily fall in line because you create an incentive structure. So the question then becomes: how does the incentive structure relate to what is already going on? And, can the new incentives be readily integrated in the games that are going on, they are totally contradictory with them? If this is so, there is going to be a big problem. If they are too easy to integrate then they do not do too much good, and if they are too different they will not be probably very well implemented. Thus, I do not really know what to do with that language because of its importance; I mean (if I use/adopt this language) I am not going to attend the limits of our power and intentions. I am always aware of the hundred thousand unanticipated consequences that any kind of voluntary design will have. In sum, I will say (that you can) design an incentive structure but the problem starts then.

MARCH: I think there are two related questions. One is the question of complexity; you are talking about relatively complex systems that are unlikely to respond to relatively simple minded manipulations, in relatively simple ways so there is the problem of complexity. The other problem is the question of the extent to which human action is based on incentives; whether behavior is interpretable in that way, if it can be either or not categorical, and if you can define incentives in such a way that any behavior is interpretable within that framework. I think, however, that we know rather well that a lot of human behavior is driven by expectation of consequences, and consequences are what you manipulate through incentives but a lot of human behavior does not fit that model very well. A lot of human behavior is much more following the rules of identity, of what I call the logic of appropriateness, and that kind of behavior does not fit incentive manipulation very well. Therefore, you have the problem of the extent to which you can manipulate incentives, the extent to which incentives control action precisely, and the extent to which incentives are not even relevant to a lot of the behavior we observe in organizations.

FRIEDBERG: I agree, I would add that even people who are responding to the incentive structure do not necessarily respond to the sense and direction that the structure provides; in other words, there are many ways in which you can use rules, directives, incentives to do things that are not quite in line with the intentions of the incentive structure's design.

MARCH: Another issue to be considered is the extent to which the incentive is approached and how it tends to drive people to behave in response. In some sense, if you design an organization with the idea that people are responsive to capable incentives, over time they would probably become so. That poses some problems for people who are concerned in things like justice and equality, which tend to not be responsive to incentive structures and they can be responsive more to identity structures.

ARELLANO: Accepting the issue of unexpected consequences or that people's behavior is not necessarily affected by incentives, the problem is that, in several countries including Mexico, several important public administration reformers are adopting this strategy as their foundation. And it is not only in government, for example we can see that universities create incentives for academics to publish in good journals, and that becomes the cornerstone to evaluate academic performance. Imagine the same in other sectors, for instance in education, where you have a proposal for a vouchers' scheme in order to create quasi–market structures. The problem is, as pointed by James March, that we have now these reforms in action and people believe that their behavior should be following incentive structures. I mean, they may be convinced that incentives are the only way through, the only manner in what an organization may obtain co–operation or make them behave in the way the organization desires. Hence, which would be the human consequence, or the organizational consequence, of people following incentives as if they were the fundamental reason for their behavior, for their action?

MARCH: That reminds me of one of the most fundamental theorems of organization theory, which is that if you want to measure behavior and reward good performance on the measures, you will corrupt the measures; the members of the organization will figure out away to score well, without doing well and that, I think, is a very general organization phenomenon, as far as there are incentives involved to measure performance. For example, measuring productivity by counting number of papers or citations may provoke that our colleagues figure out various ways to do well on those measures, without really doing any research. That is not any different from what happens in schools or when you test what happens in business firms by constructing profit–loss statements.

FRIEDBERG: I could not agree more. I think we are faced with pathological consequences of over–measuring performance in academy. I see how recruitment commissions start working now with computers and going into the file looking just how much this person is scoring, without even looking at her/his curriculum vitae. Also, I was wondering to what extent this approach of incentive structure does not really take into account what people actually do because it is not like in economies. In actual life, you combine, you mix; you know that you have to abide to some controls, to some incentives. Nevertheless, it is a hypocritical behavior, a face construction, and you go on doing what you think you have to do to do a good job. Your conception is not only about your identity but also about what is perceived as a good job and the acceptance of colleagues that look at you; in other words, there exists local control. In the end, I do not know what to do with this complete reliance on incentive structures. Being a responsible executive in any kind of domain, would mean that you have to know all incentive structures that you can design and use, but the real problem starts afterwards. How do you monitor the functioning of the incentive structure? How do you control the many dysfunctional consequences that any incentive structure comprises?

ARELLANO: Which is the new game you are creating, actually. The new game is not necessarily better or worse than its predecessor. Perhaps it is that you have created a new equilibrium and you have to deal with the good and bad consequences of its design.

Now let me ask you one question that I really would like to hear you about. I think one of the basic assumptions this vision of incentives has is a sharp distinction between organizations and institutions. With all this old, new and post–new institutionalism, what is your position regarding the difference or the relationship between this two very important concepts. I guess this is part of the problem with the idea of creating behavior through incentives.

MARCH: For me, an institution is a collection of rules and organizations can be thought as instruments for acting within rules. Organizations are also collection of rules, so I do not make a sharp distinction between institutions and organizations. I would say that some people emphasize the institutional aspects of organizations, which means to focus on stable rules and how they go over the time. However, suppose I make a distinction, what difference does it make?

FRIEDBERG: I do not see a sharp difference either. Those who make a distinction see organizations as hierarchy and institution as rules of the game. I am not comfortable with this vocabulary, but they say that, first, you change the rules of the game and, then, you will have organizations that will implement these new rules of the game. However, it is not that you have an institutional context and then you will have an organization, which is only determined by this institutional context. It is all interactive, the institutional level is influenced, impacted by what is happening on the other levels. So this how would I say? It is a hierarchical vision of society: you have all encompassing rules of the game which are ever going to change these all encompassing rules of the game? Nobody can ever change this; all we can do is try to initiate processes whereby some of these aspects will be changed.

MARCH: Intelligent powerful people have tried to make a distinction and our job is partly to try to figure out why, even though we do not see it very clearly. I think they refer to a difference in the rate of change per time. The rules of institutions certainly change over time but they do it in a slow rate compared with those rules of organizations. In that sense, I think there is a possible distinction one could make. Institutions and organizations obviously interact but the differences in the rate of change probably have consequences.

ARELLANO: I wonder whether we are dealing here with what we can call paradigms; my concern will be that paradigms are kind of "force–ideas". You try to shape reality by the concept of this paradigm and you would like reality to become the paradigm. Thus, what if we cannot make a clear distinction between institutions and organizations in academic terms, but what we really face is a kind of an ideological paradigm. In this sense, we would like to shape reality in the way we would like to see the reality to be.

FRIEDBERG: I totally agree with that these paradigms always overestimate our capacity of designing. It is very complicated to have a balance; we all know people who have been able to move organizations in a really different direction, or a country, to obtain a different climate in society. However, we just make up narratives around those persons and it is a much more complex process. Probably, this complex process requires some identification with people. In the end, if I were polemical I would say that when technocrats talk about the design of institutions, I do not really know what they are talking about.

MARCH: I consider that the concept of design applies to organically changing systems, but it does not mean that you cannot intervene in some ways to change the direction of things. Nevertheless, the design refers to something you put down on a piece of paper that is equivalent to a picture. But then, try to produce that picture (make it reality); I have never seen that happened in an organization.

FRIEDBERG: Let me just go a step further; this is exactly what I am trying to teach the students in the Master in Public Affairs. I am trying to create policy makers who are skeptical enough about their own possibility to control and design things. I tell them: "you can design whatever you want; the problem starts when you stop designing and implementing. Please take care of the process, not of the design". On the other hand, if I want to have policy makers who are willing to do things they need to design apart just to start going. How do you combine this? How do you avoid producing students who are so skeptical and so aware of the limitations of what they can do that they stop doing anything?

MARCH: It is difficult indeed, partly because we have built a society which believes that you start with the goal, with the end point, and then you construct the way to the end point. I consider that what you really want is a conception of design that does not know exactly where it is going to end. However, with certain kind of things you can initiate a movement, and perhaps —to some extent— even to provide direction to that movement. I used to contrast the attitude of a power boat skipper with a sailing boat. In the latter you cannot go wherever you desire. You try to intervene, to interpose your own movements with the wind to get you somewhere you might like to go. In contrast, in the power boat I just put the power on to go straight where I want to go. I think the power boat approach simply does not work in organizations.

ARELLANO: I remember this metaphor from you and Olsen about designing organizations is more like growing a nice garden rather than an architectural work of building.

MARCH: I refer to a modest intervention in a system, over which you have small controls. But you might have some, which I will not attribute to your own; you may not even know where you want to go. We all have experienced that our objectives change over time and the design, the intervention in the organization, requires anticipating this. How do you take the first step when you do not know where you are going? That kind of question is a difficult one but it is the kind of question you would want the students to think about.

FRIEDBERG: It is also difficult tough because it requires a high degree of tolerance to uncertainty.

MARCH: ...and the recognition that God does not create the preference function, that it evolves just as everything else does.

ARELLANO: Do you consider our students would be happy or frustrated with this kind of lesson?

MARCH: Well, perhaps you do not know exactly where you are going but you can make the way to go there beautiful. In this sense, what is really important is that if you are going to write a memorandum, it should be a literally master piece, or if you write a post notice it should be an artistic master piece; if you are going to deal with someone you should make her/him better as a result of the dealing. In every moment of your experience you should be interested in the beauty of those interactions, those hard effects you see when you are leading organizations. In this manner working along might be more interesting; I cannot derive a set of axioms from this, but I believe in it.

FRIEDBERG: It is belief that we need, which is something that makes you start things; belief and discontent with a situation, or at least your understanding of that situation. Regarding the interest of the students, I do not think that they act uniformly; some are very irritated when you tell them that there is no manner to guarantee a starting point, it is just not tolerated by some; others love it and eat it up, and a few can be convinced. Hence, for some it is natural, for others it is impossible to understand, and then you have those you can win over in a way.

ARELLANO: Would you think that in this competitive world, students would ask us good instruments, something to compete in the world, some specific instruments or procedures, the ten steps for a good organizational design? I have heard this kind of request from my students. They say: "you cannot send me to the streets without instruments; what you are giving me it is nice, interesting, but, how am I going to compete if I do not see a clear instrument?" Students will ask us to develop a clearer instrument or procedure, or vision of organization than the ones we have now.

FRIEDBERG: You have disciplines and you have skills. And the skill set I have is not one that can be put in recipes and given to students. It is a reflective skill set that can teach students to become reflective, and I think that this is an important competence for somebody who aspires to be responsible for something in the world. If he or she is not reflective he or she is not going to be a good official. I have a method but it is much less formalized and structured than statistics, which is something you go into and come out with a few formulas, things that you apply in an easy manner. In contrast, sociology and sociology organization analysis, the management of change, policy making, these are all very complicated things and there is no easy solution.

MARCH: I think very similarly. When I was younger I used to teach statistics and I did it quite deliberately because, when I taught organizations or sociology, I hope the students learned something, but it was not obvious. Conversely, when I taught statistics at the end of the time they could say "yes, I know something I did not know at the start", and that reassured me that I knew something about teaching. Now I am inclined to say to students that these things you know how to do, you have to keep doing them to maintain yourself agile because without them you cannot function properly. And so many of the organization skills that we teach are like finger exercise for a pianist: you should know how to do them, you should be able to do them if the occasion arises but they are not central. They are not the central part of your life, but you need to know all kinds of techniques in order to do something well, but the technique alone will not do it for you.

FRIEDBERG: You can organize things in such away that the students obtain an experience. I give you an example: my students of the mba go on a study trip which takes them into a big city and they have to make an analysis of a policy problem. When they came back they started a course in comparative public management and —I was amazed— after the first class a number of students came and said "this is a great course!" I got the answer of why this happened when they presented within the course the results from their study trip. A number of them said "this we analyzed in the public management course, but we have seen it". In other words, for them the topic of the course became real, they understood something and that is an experience. If you organize the teaching around some experience there will be something that can help them understand things that cannot be formalized in formulae or recipes. Therefore, it is possible, to certain extent, to organize the teaching around such moments when they say "oh yeah, I see what he means, I see it here on something I know". This is a little trick I have learned.

MARCH: I think it is very impressive, I have seen some of my colleagues do very similar things which impress me very much... I also think that occasionally you have to say we are dimming in ideas, and ideas are something that individuals have to make meaningful.



PROFESSOR: Although it is sufficient the confusion with trying to distinguish between organizations and institutions, let me introduce another element: what about systems? Institutions and organizations are systems or are very close related to the system's concept.

MARCH: Many of my colleagues are little nervous about the concept of systems because they consider that is too general. However, I think that if you use it in a disciplined way then I would say it really works. Theories of systems are about how actors are tied together in complex, independent, ways. And the whole beauty of the so–called systems analysis is feedback groups; I consider that much of the more interesting things we do in studies of organizations are when we identify feedback groups and their consequences.

FRIEDBERG: I could not agree more. I have tried to use systems a lot, but trying to get away from this cybernetic idea of the world in which there is too much control. Real systems, the things we can analyze, are never totally controlled. Therefore, I can perfectly talk about totalitarian systems but, at the same time, I am always afraid that cybernetics is associated with systems. Systems are perspectives with so many ramifications and meanings that you probably have to use metaphor to clarify what you are really talking about.

PROFESSOR: Can you identify particular trends in the study of organizations?

FRIEDBERG: I would see two trends. One is probably going into the direction of the dissolution of organizations. We call organizations to different things nowadays and we do not believe anymore in this mythical hierarchy. Another is the future of economics in organization theory, meaning that it has widened its scope to the study of markets. That is a very significant trend in society by which bureaucracy tend to dissolve in the process; dissolving meaning that it is less important for the social fabric in comparison to markets. When you compared the old organic, organized societies that we had in the fifties and sixties —well regulated, well controlled, with cooperative actors negotiating regulations— to what we have now, it is totally different. And that is reflected in what is going on in organization theory, which has become more and more economic sociology, and the fact comes from a third trend that I will add; the growing importance of network study. Network as a methodology is becoming more common, and more people use it in the study of markets but also more generally.

MARCH: I think my answer will be similar but we are moving to more complex networks, more creative processes. I reasonably have had a number of conversations with Chinese colleagues trying to understand ambiguity. Some issues, like contradiction, and how it fits into organizations, fascinate us; so, I think that quantitative studies about organizations are declining: there are more and more case studies, single case studies and efforts to develop methods. I am not sure how long a trend will continue, but I think it is interesting that over the last fifty years or so, organization studies —particularly in the United States— have moved increasingly toward business schools. In contrast, the political science contribution to organization studies, has declined, also psychological contributions have declined. Conversely, sociological contributions have even increased, but the most obvious structural thing is the movement to business schools and therefore the increase in emphasis upon business and particularly from the point of view of the firm, rather than the perspective of the industry or the society. The foolishness of predicting trends is what we do to extrapolate current trends and why they are interesting changes and trends. I do not really know what changes there will be in those trends, but I can perceive the obvious things are going on.

ARELLANO: This is a very tough question. I have been in the editorial board of Organizational Studies for the last three years now, and every week I receive three or four papers to review. My sense is that I agree with Jim, business schools are dominating heavily, at least in Organizational Studies. Another thing that it is very important for us, here at CIDE and also in the Autonomous Metropolitan University, is that public organizational studies are now a minority in organizational studies. We will have a workshop, next May in France, organized by Organizational Studies and the topic will be to bring the state back in organizational studies. It is very important that professors and students in organization do make studies about the public sector, especially now that the current issue is analyzing organizations as firms. There are very few good studies about public organizations and even less in other kind of organizations such as the judiciary or international organizations. Therefore, I would say that one important trend will recover the study of the so called (I quote) other organizations: public organizations, international organizations, judicial organizations, rather than just analyzing firms.

FRIEDBERG: I would only add that my concern is not that business schools are becoming a major actor, but that they have been a major actor for long time in organizations studies. Rather, from my perspective I would be very glad if they really took seriously the study of business organizations. The problem is that they do not study them properly and, therefore, the knowledge they produce about business is superficial, which is really dangerous.


About the authors

* James G. March is Professor Emeritus at Stanford University, where he has been on the faculty since 1970. He holds appointments in the Schools of Business and Education and in the Departments of Political Science and Sociology. He is best known professionally for his writings on decision making and organizations, including Organizations, A Behavioral Theory of the Firm, Leadership and Ambiguity, Ambiguity and Choice in Organizations, Decisions and Organizations, Rediscovering Institutions, The Pursuit of Organizational Intelligence, Democratic Governance, A Primer on Decision Making, The Dynamics of Rules, Explorations in Organizations, On Leadership, and The Ambiguities of Experience.

** Erhard Friedberg is Austrian by birth and education. His career has been conducted largely in France where he has completed numerous empirical studies in a variety of settings both in private firms, and especially the French automotive industry, and public policy arenas such as the implementation of industrial and cultural policy in France, or the comparison between the French and the German university systems. Professor Friedberg is both a researcher and a teacher. He presently directs the Master of Public Affairs at Sciences PO and the Post–Graduate Program in Sociology at Sciences PO (the Paris Institute of Political Science). He is a frequent consultant to Renault and other French or European Firms, as well as the author of numerous articles on the theory and practice of organized action.

*** David Arellano Gault is professor and researcher in Centro de Investigación y Docencia Económicas (CIDE). He is doctor of Public Administration by Colorado University. Currently he is editor of Gestión y Política Pública, co–editor of Organization Studies and member of the editorial board oí Public Administration Review. He is also member of the National Researchers System (SNI), level III. He has published several books on public management and more than 50 articles both in national and international magazines.

Creative Commons License Todo o conteúdo deste periódico, exceto onde está identificado, está licenciado sob uma Licença Creative Commons