Content uploaded by Sven Steinmo
Author content
All content in this area was uploaded by Sven Steinmo on Nov 12, 2019
Content may be subject to copyright.
Institutionalism
Sven Steinmo
European University Institute, Florence, Italy
June 27, 2013
The study of political institutions has moved back into center stage in political science.
Whereas only a few years ago, institutions were mostly casually mentioned in most political
science research, today they are a central focus of attention. This change is not simply a
change of language or the catching on of a new popular phrase or academic fad. Instead these
changes represent an important development in evolution of political science theory and
intellectual focus.
The following essay surveys the history of institutionalist thought, explores several different
types of institutionalist theory and suggest the central analytical agendas of modern
institutionalist theory.
1 What are Institutions?
In the broadest sense, institutions are simply rules. As such, they are a foundation for all
political behavior. Some are formal (as in constitutional rules) some are informal (as in
cultural norms), but without institutions there could be no organized politics. Indeed absent
institutions there could be no organization at all. To understand this point, simply attempt to
consider a world in which there were no rules: In this Hobbesian hell, individuals would be
forced to ‘invent’ communication every time they encountered another individual. In this
sense, then, if we study social interaction, we study institutions. This does not suggest,
however, that all social scientists are ‘Institutionalists.’
2 Institutionalism
The ‘Institutionalism’ specifically examines the ways in which institutions structure social
and political behavior (North 1990). This burgeoning body of literature argues that policy,
politics, and behavior can only be understood in the context of the institutions in which they
take place. Thus, for example, Ellen Immergut argues that variations in National Health
Insurance (NHI) systems are best explained by variations in national political institutions
(Immergut 1998). Similarly, Bo Rothstein shows that Sweden's high union density is best
explained by the ‘Ghent’ unemployment insurance system which gives workers powerful
incentives to join Swedish unions (Rothstein 1992). Even more broadly, Douglas North
suggests that the very success of Western political economic model is rooted in the peculiar
institutions developed in these societies (North 1990).
The central tenet of this New Institutionalism is that institutions are not neutral to policy
outcomes. As Peter Hall has suggested, ‘On the one hand, the organization of policy-making
affects the degree of power that any one set of actors has over policy outcomes…. On the
other hand, organizational position also influences an actor's definition of his own
interests…in this way, organizational factors affect both the degree of pressure an actor can
bring to bear on policy and the likely direction of that pressure’ (Hall 1986) p. 12..
In sum, institutions define the rules of the political game and as such they define who can
play and how they play. Consequently, they ultimately can shape who wins and who loses. If
politics is the study of who gets what, when and why, then institutionalists argue that
institutions should be at the heart of that study.
3 A Brief History of Institutionalism
Institutionalism has a long established tradition among those interested in politics and
political outcomes. Plato's Republic is a comparative study of institutions. Similarly,
Aristotle's central concern in Politics is which kinds of political institutions will produce the
best outcomes. James Madison must clearly also be seen as an early American
‘Institutionalist’ in that he was specifically concerned with which kinds of institutions would
produce the best political outcomes and how the specific design of institutions would shape
political outcomes.
Political scientists have also long been interested in institutions. Indeed, in its early years
political science meant the study of political institutions (Wilson 1891). But, with some
important exceptions (Herring 1940; Key 1947), early political science was often more
descriptive than analytical. ‘Comparative politics,’ in particular, consisted mostly of detailed
configurative studies of different legal, administrative, and political structures. In the
immediate postwar years a new generation of political scientists attempted to make the study
of politics more ‘scientific.’ For many this effectively meant that political science ought to
model itself on the ‘hard sciences’ which they believed was fundamentally a deductive
process. Thus rather than studying the details of political life and inductively uncover the
patterns of behavior and action, political ‘science’ should be a deductive science that seeks to
discover of the general laws and fundamental forces that lay behind political action. Focusing
on particular institutions, proponents this intellectual agenda implied, was ‘a-theoretical.’
Concomitant with the push for more abstract laws, political scientists were disillusioned by
the failure of parliamentary institutions in Inter-war Weimar Germany (and later in post-
colonial Africa) to prevent these polities from devolving into authoritarianism. Clearly, many
argued, there were bigger, more important, indeed more fundamental, forces at work in
politics and development than political institutions. These forces, they argued, should be the
focus of political science.
Thus, behavioralist, functionalist and Marxist perspectives took leading roles in political
science theory building through most of the 1960s and 1970s. As a consequence, institutional
analysis diminished in prominence…particularly in comparative politics. For Functionalists,
political institutions were simply seen as organizations created to fulfill the systemic needs of
society. As such, political institutions were neither important nor interesting. For
Behavioralists and Marxists, political institutions were simply arenas in which political battles
were fought out. While Marxists saw the relevant groups as classes and Pluralists saw them as
more narrowly defined, neither theoretical perspective played particular attention to the
structure or character of the institutions themselves. The real meat of politics, they argued—
and the keys to understanding political outcomes—was found in the articulation of group
interests. Though this was rarely explicitly stated, implicit in these theories was the assertion
that if politics or policies differed between societies, this difference was surely the result of
different constellations of group and class interests or preferences (Almond and Verba 1963;
Verba 1967; Miliband 1969).
It was not the case, however, that all political scientists had abandoned the study of
institutions. Indeed, many of the most widely read scholars in American Politics maintained
an explicitly institutional emphasis (Schattschnieder 1960) (Greenstein 1963). In
Comparative Politics, as well, several political scientists continued to examine political
institutions and their effects on political outcomes (Ekstein 1960; Bendix 1964; Huntington
1968), even while they sometimes had to defend their ‘inductive’ approach from the challenge
that it was not ‘scientific’ (Przeworski and Teune 1970). For these scholars it was self-evident
that if one wanted to understand what government does, one needs to specifically study the
institutions through which it acts.
It was probably in Comparative Politics that the search for ‘Grand Theory’ had its most
significant impact. It may be for this reason that a self-conscious return to ‘Institutionalism’
was most forcefully articulated here. Among the first group of scholars to move in this
direction was Peter Katzenstein and his colleagues who sought to explain why several
advanced capitalist states responded so differently to the oil shock of the 1970s. The answer,
they concluded, was found in the differing institutional structures in these polities and the
consequent patterns of economic policy pursued in each nation (Katzenstein 1978).
Similarly, Theda Skocpol's study of social revolutions also concluded that one could not
explain the course of a country's revolution without examining the nature and structure of the
state against at which these revolutions were aimed (Skocpol 1979).
Soon a new ‘state centered’ approach emerged in comparative political inquiry. These
scholars forcefully argued against the behavioralist and Marxist ‘grand theory’ emphasis then
dominant in the study of comparative politics. Instead, they suggested, political outcomes
were shaped and structured by the specific actors and their position in the state. One should
not treat the state as a neutral ‘black box’ through which group or class interest was simply
translated (Evans, Rueschemeyer et al. 1985). Instead, the state had independent interests and
agendas which were separate and different from the interests and preferences of classes and
interest groups that made up society.
Of course it did not take long for these scholars to discover that ‘the state’ is too broad a
concept and too varied a set of institutions to be ‘taken seriously’ without being broken down.
To make analytic sense out of the insights, scholars began examining institutions more
carefully. If state institutions ‘matter,’ they argued, then why should this not be equally true
of institutions outside the boundaries of the executive state? With these questions ‘The New
Institutionalism’ was born.
4 Three Types of Institutionalism
Today three different intellectual approaches lay claim to the term ‘Institutionalist’ (Hall and
Taylor 1996). Each grows out of a different academic discipline and attempts to integrate
these different analytic traditions into the understanding of politics. Sociological
Institutionalism, as its name implies, grows out of sociology and the study of organizations
(Selznick 1949). These scholars have been centrally interested in understanding culture and
norms, as institutions. These scholars emphasize ‘folkways,’ ‘patterns of behavior,’ and
‘cognitive maps’ and argue that these social institutions are critical for understanding the
structure of social, political, and economic interactions (March and Olsen 1989; DiMaggio
and Powell 1991). Building on upon their analyses of complex organizations, these scholars
show the relationship between formal institutions and the structure or patterns of behavior and
beliefs. They argue that these informal institutions are core to any understanding of the
nonrational aspects human communication and exchange.
‘Rational Choice Institutionalism,’ taking its lead from economics, emphasizes quite different
sets of institutions and their effects. Rational Choice (RC) scholars attempt to apply the
formal logic and methods to the study of politics and history and often explicitly eschew the
more ‘mushy’ or less precise variables such as norms and beliefs. Instead, basic assumptions
are made about human behavior and motivation in order to uncover and examine the basic
laws of political behavior and action. Scholars in this tradition argue that once these laws are
discovered, models can be constructed that will help us understand and predict political
behavior (Levi 1988). In their deductive model, Rational Choice scholars look to the real
world to see if their model is right (test the model). For these scholars, understanding real
outcomes is not the first point—creating, elaborating, refining a theory of politics is
(Weingast 1996).
The implications of this scientific orientation are substantial. Morris Fiorina, a highly
regarded RC scholar at Harvard, put the issue in the following way: ‘most PTI scholars are
not as interested in a comprehensive understanding of some real institution or historical
phenomenon, so much as in a deeper understanding of some theoretical principle or
logic…[F]or most PTI scholars, breadth trumps depth; understanding 90 percent of the
variance in one case is not as significant an achievement as understanding 10 percent of each
of nine cases, especially if the cases vary across time and place’ (Fiorina 1995) p. 110-1.
The third ‘New Institutionalist’ approach emerges out of what might be considered a more
traditional political science. Consequently, it has a quite different aim: Historical
Institutionalists are primarily interested in understanding and explaining real world events and
outcomes. As with the other approaches noted above, scholars working in this tradition also
argue that one cannot explain particular historical outcomes without specifically examining
the way in which the political institutions have shaped or structured the political process
(Steinmo, Thelen et al. 1992). But, unlike RC scholars in particular, Historical Institutionalists
came about their ‘institutional’ arguments inductively after testing a variety of alternative
theories (i.e., Marxist, structural functionalist, culturalist, and rationalist) against the
outcomes they observed. In other words, Historical Institutionalists are first interested in
explaining an outcome (say, for example, why France and Britain have pursued such different
styles of industrial Policy (Hall 1986) or why some welfare states generate more popular
support than others (Rothstein 1998); they then proceed to explore alternative explanations
for the outcomes they observe.
Historical Institutionalists do not argue that institutions are the only important variables for
understanding political outcomes. Quite the contrary, these scholars generally see institutions
as intervening variables (or structuring variables) through which battles over interest, ideas,
and power are fought. Institutions are thus the focal points in critical junctures in a historical
path analysis because political battles are fought inside institutions and over the design of
future institutions.
5. Taking History Seriously
These insights had important implications, both for what we study and for how we study it.
Historical institutionalists study history because they believe history matters, not merely to
increase the reference points for analysis (as is done in time series analysis). There are at least
three important ways in which history matters. First, political events happen within a
historical context, which has a direct consequence on the decisions or events. An early
example of this is the seminal work of Alexander Gershenkron, who argued that when a
country industrializes necessarily affects how it industrializes. He shows us why latecomers
cannot go through the same long trial-and-error process followed by early developers.1 In
other words, the process of industrialization is essentially different for late developers than for
early developers. This is a huge insight that is easily missed in large scale quantitative, cross
national comparisons, which very often pool data across continents and time periods and treat
the time/place as inconsequential (or assume that it will ‘wash out’ of the analysis).
The second reason history matters is that actors or agents can learn from experience.
Historical institutionalists understand that behaviour, attitudes and strategic choices take place
inside particular social, political, economic and even cultural contexts. Rather than treating all
political action as if fundamentally the same irrespective of time, place or context, historical
institutionalists explicitly and intentionally attempt to situate their variables in the appropriate
context. Thus, by deepening and enriching their understanding of the historical moment and
the actors within it, they are able to offer more accurate explanations for the specific events
that they explore than had they treated their variables outside the temporal dimension.
E.E. Schattschneider’s early work on tariff policy showed how political choices made
1 An example outside politics may prove illustrative. Many of us recognize that firstborn
children have a very different developmental experience than second (or later) children. Not
only are the parents more experienced after the first child, they are also taking care of more
than one child at a time. Finally, and equally importantly, subsequent children grow up in a
home where there are older siblings – something the first child, by definition, cannot do.
at time A have important consequences for time B. In this work he famously argued that ‘new
policies create new politics’. Following Schattschneider, Paul Pierson has shown in several
important works how and why policy choices at one point in time affect choices at subsequent
points in time. Similarly, Esping-Anderson pointed out in his seminal Three Worlds of
Welfare Capitalism how, given the fact that we live in modern welfare states with
unemployment insurance, health insurance, pension programs and the like: The existence of
the welfare state is a fact of modern political life that itself shapes politics, expectations and
policy in the countries that have developed it (Esping-Andersen 1990).
Finally, again as Pierson has shown, expectations are also moulded by the past
(Pierson 2004). While some might point to America’s adventure in Iraq as a simple product of
power politics and/or the demand for oil, a historical institutionalist would more likely look to
the patterns of past victories for an understanding of why this country reacted in the way it did
to the 9/11 bombings. Certainly they were mistaken, but there should be little doubt that
America’s past successes in Germany and Japan – to say nothing of their perceived victory
over Communism at century’s end – led policy makers in the administration to believe that
they could assert American power and bring successful capitalism and democracy to a former
dictatorship.
In sum, for historical institutionalists, history is not a chain of independent events.
There is more than the temporal dimension implied in this basic point. Taking history
seriously ultimately means that the scholar is sceptical of the very notion of variable
independence. Instead, acknowledging the importance of history suggests an explicit
awareness that important variables can and often do shape one another. Historical
institutionalists, more than political scientists in some other traditions, are explicitly interested
in these interactive effects on the interdependence of multiple causal variables.
Historical institutionalists are something like the environmental biologist who believes
that in order to understand the specific fate of a particular organism or behaviour, she must
explicitly examine that organism or behaviour in the ecology or context in which it lives. This
implies a different scientific ontology than that commonly found in the hard sciences of
physics and chemistry. At the root of evolutionary biology is the assumption that the objects
of analysis – living organisms – are fundamentally different from inanimate matter. While
objects in the physical world often adhere to constant ‘laws’ of nature, biological organisms
often defy attempts to reduce them to their essential components because of their complexity.
Thus, as eminent evolutionary biologist Ernst Mayer points out, the development of biology
as a science has required an investigation of ‘additional principles’ that applied only to living
organisms. He argues, ‘This required a restructuring of the conceptual world of science that
was far more fundamental than anyone had imagined at the time’ (Mayr 2004) p. 26.
Historical institutionalism represents something like this ontological move in social
science. In order to understand historically specific events and long term political outcomes,
one could not strictly apply methods and epistemologies drawn from the study of invariant
variables that have fixed relationships across space and time. This, of courses, does not mean
that it is not science – unless one’s definition of science would exclude biology as well;
rather, it implies that the scientific methods applied should fit the subject being studied.
6. Ideas, Institutions and Change
To argue that “institutions matter” is by now taken as common sense. Today it is
widely recognized that institutions provide stability in political life because they structure
political choices. This is because when we know the rules, we are more likely to know how to
interact and how to behave. Indeed, when we know the rules we adapt our behavior
accordingly. The more difficult and interesting questions arise when we ask: 1) Why do
people follow rules? 2) How do we explain institutional change? And, 3) if institutions are so
important, what role is left for human agency and ideas?
Early institutionalist scholars drew insights from evolutionary theory of "punctuated
equilibrium" to explain institutional change (Steinmo, Thelen et al. 1992). In recent years,
however, there has been growing interest in explaining the endogenous sources of
institutional change. A number of scholars have mapped out different types of institutional
change in different contexts (Streeck and Thelen 2005; Mahoney and Thelen 2009). These
efforts have led to considerable refinement of our understanding of different types of
institutional change, but are insufficient for explaining why institutions change.
The most important contributions to Institutionalist theorizing on change have come
from those scholars who specifically have tried to examine relationship between human
agency and ideas (Campbell 2002; Lieberman 2002; Blyth 2008; North 2008). Traditionally,
(especially those in the rational choice school) institutions are created by individuals seeking
maximize their individual interests. Ideas, in this view, were simply epiphenomenal or
justification for people’s ‘real’ motivation. But such an approach is unsatisfactory; both at the
common sense level (we know that human beliefs and ideas matter for our choices) and
because these approaches fail to explain how we move from one equilibrium to another.
The interesting questions arise when we look at the interdependent relationships
between ideas, institutions and interests over time. Since none of these factors are static, then
only through an historical analysis can we learn how they develop, how they change and why
they are so different in different contexts (Lieberman 2002). Building on the work of social
psychologist such as Amos Tversky and Daniel Kahneman (Kahneman, Slovic et al. 1982;
Kahneman 2011), institutionalist scholarship is moving toward a more nuanced understanding
of the ‘social foundations’ of human rule following behavior and are attempting to better
understand the ways in which human cognitive biases’ shape interact with institutional
structures to influence human behavior. Social psychologists have long understood that
human beings are social creatures who are not normally individualistic in a meaningful sense.
Instead we are driven to follow social rules and look for social approval. Human beings are
more likely to be normative actors, possessing significant cognitive biases,’ than rational self-
interested utility maximizers. Appreciating human motivations in this way forces us to
consider the interaction of institutional rules, human choices and eventually, differences in
cultural systems.
Finally, institutionalists have begun to argue that we can understand change in
evolutionary terms and thus bring some of the insights from evolutionary theory into the
study of institutional evolution (Lewis and Steinmo 2012). Whereas traditional political
science has taken a mechanical approach to the study of politics and human affairs (Hall
2003) an evolutionary approach to institutional change allows one to integrate contingency
into the study as well be more honest with the fact that there are no truly independent
variables of interest in human affairs . Almost everything that is really interesting and
important is instead the product of the complex interaction of multiple factors that interact
with each other over time (Streeck 2010).
See also:
Institutionalization; Institutions; Law: New Institutionalism; Norms; Political Science:
Overview; Political Sociology; Rational Choice in Politics
Bibliography
4314 words
References
Almond, G. A. and S. Verba (1963). The civic culture; political attitudes and democracy in
five nations. Princeton, Princeton University Press.
Bendix, R. (1964). Nation-building and citizenship. Berkeley, University of California Press.
Blyth, M. (2008). ELEN degenerates: Four (Plus Two) reasons to Take Ideas Very Seriously
Indeed. Ideas and Politics in Social Sciences. D. Beland and R. Cox, forthcoming.
Campbell, J. (2002). "Ideas, politics and public policy." Annual Review of Sociology 28: 21-
38.
DiMaggio, P. J. and W. W. Powell (1991). Introduction. The new institutionalism in
organizational analysis. W. W. Powell and P. J. DiMaggio. Chicago, The University of
Chicago Press: 1-38.
Ekstein, H. (1960). Pressure group politics: the case of the medical association. Stanford, CA,
Stanford University Press.
Esping-Andersen, G. s. (1990). The three worlds of welfare capitalism. Cambridge, Polity.
Evans, P., D. Rueschemeyer, et al. (1985). Bringing The State Back In. New York,
Cambridge University Press.
Fiorina, M. (1995). "Rational Choice and the New (?) Institutionalism." Polity XXXVIII(1
(Fall)): 107-115.
Greenstein, F. I. (1963). The American party system and the American people. Englewood
Cliffs, Prentice-Hall.
Hall, P. (1986). Governing the Economy. New York, Oxford University Press.
Hall, P. (2003). Aligning Ontology and Methodology in Comparative Research. Comparative
historical analysis in the social sciences. J. Mahoney and D. Rueschmeyer. Cambridge,
Cambridge University Press: 373-404.
Hall, P. and R. Taylor (1996). "Political science and the three new institutionalisms." Political
Studies 44: 936-957.
Herring, P. (1940). The politics of democracy; American parties in action. New York, W. W.
Norton & Company.
Huntington, S. (1968). Political Order in Changing Societies. New Haven, Yale University
Press.
Immergut, E. (1998). "The Theoretical Core of the New Institutionalism." Politics and Society
26(1): 5-34.
Kahneman, D. (2011). Thinking, Fast and Slow. London, Penguin.
Kahneman, D., P. Slovic, et al., Eds. (1982). Judgement under uncertainty: Heuristics and
biases. New York, Cambridge University Press.
Katzenstein, P. J. (1978). Between power and plenty : foreign economic policies of advanced
industrial states. Madison, University of Wisconsin Press.
Key, V. O. (1947). Politics, parties and pressure groups. New York, Thomas Y. Crowell
company.
Levi, M. (1988). Of rule and revenue. Berkeley, University of California Press.
Lewis, O. and S. Steinmo (2012). "How Institutions Evolve: Evolutionary Theory and
Institutional Change." Polity 44(3): 314-339.
Lieberman, R. (2002). Political institutions and the politics of race in the development of the
modern welfare state. Restructuring the Welfare State-Political Institutions and Policy
Change-. B. Rothstein and S. Steinmo. New York, Palgrave Macmillan: 102-129.
Lieberman, R. C. (2002). "Ideas, Institutions, and Political Order: Explaining Political
Change." The American Political Science Review 96(4): 697.
Mahoney, J. and K. Thelen (2009). How Historical Institutionalists explain change.
Explaining Institutional Change: Ambiguity, Agency, and Power in Historical
Institutionalism. J. Mahoney and K. Thelen. New York, Cambridge University Press.
March, J. and J. Olsen (1989). Rediscovering Institutions. New York, Free Press.
Mayr, E. (2004). What makes biology unique? : considerations on the autonomy of a
scientific discipline. Cambridge, UK ; New York, Cambridge University Press.
Miliband, R. (1969). The state in capitalist society. London, Weidenfeld & Nicolson.
North, D. (1990). Institutions, institutional change and economic performance. Cambridge,
Cambridge University Press.
North, D. C. (2008). What is Missing from Political Economy. The Oxford handbook of
political economy. B. R. Weingast and D. A. Wittman. Oxford, Oxford University Press:
1003-1009.
Pierson, P. (2004). Politics in time: History, institutions, and social analysis. Princeton,
Princeton University Press.
Przeworski, A. and H. Teune (1970). The logic of comparative social inquiry. New York,,
Wiley-Interscience.
Rothstein, B. (1992). Labor market institutions and working class strength. Structuring
Politics: Historical institutionalism in comparative politics. S. Steinmo, K. Thelen and F.
Longstreth. New York, Cambridge University Press: 33-56.
Rothstein, B. (1998). Just Institutions Matter: The Moral and Political Logic of the Universal
Welfare State. New York, Cambridge University Press.
Schattschnieder, E. E. (1960). The semi-sovereign people. New York, Holt Reinhart.
Selznick, P. (1949). TVA and the grass roots : a study in the sociology of formal organization.
Berkeley, University of California Press.
Skocpol, T. (1979). States and Social Revolutions: A Comparative Analysis of France,
Russia and China. Cambridge, Cambridge University Press.
Steinmo, S., K. A. Thelen, et al. (1992). Structuring politics : historical institutionalism in
comparative analysis. Cambridge [England] ; New York, Cambridge University Press.
Streeck, W. (2010). Epilogue. Institutions in History: Bringing Capitalism Back In. Handbook
of Comparative Institutional Analysis, , Ox-ford: Oxford University Press, 2010. J. Campbell,
C. Crouch, P. H. Kristensenet al. Oxford, Oxford University Press.
Streeck, W. and K. A. Thelen (2005). Introduction: Institutional change in advanced political
economies. Beyond Continuity. W. Streeck and K. Thelen. Oxford, Oxford University Press:
1-39.
Verba, S. (1967). "Some Dilemmas in Comparative Research." World Politics 20: 111-127.
Weingast, B. (1996). Political Institutions: Rational Choice Perspectives. A New Handbook
of Political Science. H. D. K. R.E. Goodin. Oxford, Oxford University Press: 167-190.
Wilson, W. (1891). Congressional government : a study in American politics. Boston,
Houghton.
