sense, states created markets, for the fundamentals of market exchange were only
possible in a juridical framework of commercial law created by states. Polanyi puts it
pithily: ‘‘Regulation and markets, in eVect, grew up together’’ ( 1944 – 57 , 68 ).
The ‘‘new institutionalism’’ in economics is associated with the work of North
(North and Thomas 1973 ; North 1991 ). It is striking how far this new economic
institutionalism parallels the concerns of the new institutionalism that swept over
political science following the publication of March and Olsen’s landmark works
( 1984 , 1989 ). Four similarities merit emphasis.
TheWrst is the extent to which an almost theological debate developed about the
meaning of ‘‘institutions.’’ Indeed North on institutions sounds almost mystical:
‘‘We cannot see, feel, touch, or even measure institutions; they are constructs of the
human mind’’ ( 1991 , 107 ).
In North, however, this insistence is connected to a second theme which parallels
the political science new institutionalism: The importance of distinguishing insti-
tutions from organizations. The distinction is critical because: ‘‘Institutions...
determine the opportunities in a society. Organizations are created to take advan-
tage of those opportunities’’ ( 1991 , 7 ).
Why do these opportunities exist? Because of a third feature which parallels one of
the key elements of political science institutionalism—perhaps the most
important parallel. These are the linked characteristics of feedback and lock-in.
Feedback is the process by which institutions adapt in the light of messages arising
from their preceding activities, and interaction with their environment. ‘‘Lock-in’’ is
the process by which institutions are constrained into particular patterns of devel-
opment and behavior by the impact of past actions and commitments (North
1991 , 7 ). The idea is plainly central to the wider, and more familiar, notion of ‘‘path
dependency.’’ The emphasis on ‘‘path dependency’’ turns out to have large implica-
tions for understanding change in institutional life, and for making sense of the role
of human agency in change. Of course this is to put things only in terms of the
restriction that path dependency creates. The wider literature on institutionalism
reminds us that the other side of the path dependency coin is beneWcial: It creates
routine, certainty, and trust in economic and other social exchanges (Pierson 2000 ).
How does institutional choice work? This is the fourth parallel theme uniting the
concerns of economic and ‘‘political science’’ institutionalism, and it can be
illustrated from a recurrent problem—that of understanding the signiWcance of a
peculiarly important organization, theWrm. As Moe puts it:
The neoclassical theory of theWrm is not in any meaningful sense a theory of economic
organization. It centers around the entrepreneur, a hypothetical individual who, by assump-
tion, makes all the decisions for theWrm and is endowed with a range of idealized properties
deWning his knowledge, goals, computational skills, and transaction costs. (Moe 1984 , 740 )
That problematic quality has been made more acute by the development of the
Wrm in the modern industrial economy—by the extent to which it has become, in
146 michael moran