Increasing Returns,Path Dependence,and the Study of Politics TOR Paul Pierson The American Political Science Review,Vol.94,No.2.(Jun.,2000),pp.251-267. Stable URL: http://links.jstor.org/sici?sici=0003-0554%28200006%2994%3A2%3C251%3AIRPDAT93E2.0.CO3B2-Z The American Political Science Review is currently published by American Political Science Association. Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use,available at http://www.jstor.org/about/terms.html.JSTOR's Terms and Conditions of Use provides,in part,that unless you have obtained prior permission,you may not download an entire issue of a journal or multiple copies of articles,and you may use content in the JSTOR archive only for your personal,non-commercial use. Please contact the publisher regarding any further use of this work.Publisher contact information may be obtained at http://www.jstor.org/journals/apsa.html. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is an independent not-for-profit organization dedicated to creating and preserving a digital archive of scholarly journals.For more information regarding JSTOR,please contact support@jstor.org. http://www.jstor.org/ Sun Dec1013:51:422006
American Political Science Review Vol.94,No.2 June 2000 Increasing Returns,Path Dependence,and the Study of Politics PAUL PIERSON Harard University t is increasingly common for social scientists to describe political processes as "path dependent.The concept,however,is often employed without careful elaboration.This article conceptualizes path dependence as a social process grounded in a dynamic of "increasing returns."Reviewing recent literature in economics and suggesting extensions to the world of politics,the article demonstrates that increasing returs processes are likely to be prevalent,and that good analytical foundations exist for exploring their causes and consequences.The investigation of increasing returns can provide a more rigorous framework for developing some of the key claims of recent scholarship in historical institutionalism:Specific patterns of timing and sequence matter;a wide range of social outcomes may be possible;large consequences may result from relatively small or contingent events;particular courses of action,once introduced,can be almost impossible to reverse;and consequently,political development is punctuated by critical moments or junctures that shape the basic contours of social life. t is increasingly common for social scientists to that confront them.If path dependence arguments are describe political processes as "path dependent." indeed appropriate in substantial areas of political life, Claims of path dependence have figured in both they will shake many subfields of political inquiry.This classic works of comparative politics,such as Lipset essay argues that they are. and Rokkan's (1967)analysis of European party sys- The analysis begins with a general discussion of path tems,and more recent analyses on topics such as labor dependence that seeks to clarify some important am- incorporation in Latin America (Collier and Collier biguities surrounding the concept.I then outline and 1991),the outcome of state-building processes in Eu- investigate the distinctive characteristics of social pro- rope(Ertman 1996),and the comparative development cesses subject to what economists call "increasing of health care systems (Hacker 1998).The notion of returns,"which could also be described as self-reinforc- path dependence is generally used to support a few key ing or positive feedback processes.For some theorists, claims:Specific patterns of timing and sequence mat- increasing returns are the source of path dependence; ter;starting from similar conditions,a wide range of for others,they typify only one form of path depen- social outcomes may be possible;large consequences dence.The focus here is on increasing returns pro- may result from relatively "small"or contingent events; cesses,both because they are of great social signifi- particular courses of action,once introduced,can be cance and because (in contrast to broader conceptions virtually impossible to reverse;and consequently,po- of path dependence)social scientists are beginning to litical development is often punctuated by critical develop rigorous arguments about the causes and moments or junctures that shape the basic contours of consequences of increasing returns.Increasing returns social life (Collier and Collier 1991;Ikenberry 1994; dynamics capture two key elements central to most Krasner 1989).All these features stand in sharp con- analysts'intuitive sense of path dependence.First,they trast to prominent modes of argument and explanation pinpoint how the costs of switching from one alterna- in political science,which attribute "large"outcomes to tive to another will,in certain social contexts,increase "large"causes and emphasize the prevalence of markedly over time.Second,and related,they draw unique,predictable political outcomes,the irrelevance attention to issues of timing and sequence,distinguish- of timing and sequence,and the capacity of rational ing formative moments or conjunctures from the peri- actors to design and implement optimal solutions ods that reinforce divergent paths.In an increasing (given their resources and constraints)to the problems returns process,it is not only a question of what happens but also of when it happens.Issues of tempo- Paul Pierson is Professor of Government,Harvard University,Cam- rality are at the heart of the analysis bridge,MA 02138. The following section reviews the development of An carlier version of this article was presented as the Jean Monnet Visiting Professor Lecture,European University Institute (EUI), increasing returns arguments in the social science April 1997.I am very grateful to Andy Rudalevige and Jeremy discipline in which they have received the greatest Weinstein for research assistance on this project and to the German attention:economics.This review suggests the wide Marshall Foundation,Simon R.Guggenheim Poundation,and the sweep of potential applications,even in a field that Weatherhead Center for International Affairs for financial support.I might be expected to be hostile to the idea.More have received helpful comments on earlier drafts from many people including several thoughtful reviewers for the APSR.I owe particular important,these economic applications provide the thanks to Frank Baumgartner,David Collier,Pepper Culpepper, most analytically developed discussions of increasing Jeffry Frieden,Daniel Goldhagen,Jacob Hacker,Peter Hall,Alan returns.Economists not only have clarified the princi- Jacobs,Margaret Levi,Robert Mickey,Andrew Moravcsik,Gerardo pal implications of path dependence but also have Munck,Alec Stone,and Kathleen Thelen.I also benefited from discussions at the Harvard Center for European Studies,Columbia identified many of the specific aspects of a particular University,the EUL,the Centro de Investigacion y Docencia Eco- social environment that generate such processes. nomicas (CIDE),the University of California at Berkeley,the The discussion of economics prepares the way for an University of Washington,and Yale University. exploration of the distinctive characteristics of politics. 251
Increasing Returns,Path Dependence,and the Study of Politics June 2000 Rather than simply apply extant arguments in econom- we cannot understand the significance of a particular ics to political phenomena,we need to consider the social variable without understanding "how it got features of the political world that require modifica- there"-the path it took.Previous events in a sequence tions in the use of path dependence claims.I will influence outcomes and trajectories but not necessarily demonstrate that increasing returns arguments are at by inducing further movement in the same direction. least as relevant to an understanding of politics as they Indeed,the path may matter precisely because it tends are in other areas of the social sciences.Indeed,factors to provoke a reaction in some other direction. such as the prominence of collective activity in politics, An alternative,narrower definition has been sug- the central role of formal,change-resistant institutions, gested by Margaret Levi (1997,28): the possibilities for employing political authority to magnify power asymmetries,and the great ambiguity of Path dependence has to mean,if it is to mean anything. many political processes and outcomes make this a that once a country or region has started down a track,the costs of reversal are very high.There will be other choice domain of social life that is especially prone to increas- points,but the entrenchments of certain institutional ing returns processes. arrangements obstruct an easy reversal of the initial The final section considers what these arguments can choice.Perhaps the better metaphor is a tree,rather than contribute to political analysis.They provide an impor- a path.From the same trunk,there are many different tant caution against a too easy conclusion of the branches and smaller branches.Although it is possible to inevitability,"naturalness,"or functionality of observed turn around or to clamber from one to the other-and outcomes.Given the ubiquity of claims about efficient essential if the chosen branch dies-the branch on which a or functional elements in politics,this alone would be climber begins is the one she tends to follow. an important corrective.More significant,increasing This conception of path dependence,in which preced returns arguments justify efforts to stretch the temporal ing steps in a particular direction induce further move- horizons of political analysis.They can redirect the questions political scientists ask,which will contribute ment in the same direction,is well captured by the idea of increasing returns.In an increasing returns process, to a richer appreciation of the centrality of historical the probability of further steps along the same path processes in generating variation in political life.They increases with each move down that path.This is can also direct attention toward particular variables because the relative benefits of the current activity and suggest promising hypotheses about the sources of compared with other possible options increase over both political stability and political change in certain time.To put it a different way,the costs of exit-of common political settings.For instance,increasing switching to some previously plausible alternative- returns arguments highlight the need to consider hy- rise.Increasing returns processes can also be described potheses based on temporal ordering--the possibility as self-reinforcing or positive feedback processes. that the particular sequencing of events or processes We face,then,a choice between a broader and a may be a key part of the explanation for divergent narrower conception of path dependence.Frequently, outcomes.Finally,grasping the implications of wide- authors are not explicit about which of these two spread path dependence can help orient political sci- meanings they intend.To establish greater clarity,this entists to a realistic,which is to say modest,set of essay employs the term in the narrower sense,that is, aspirations regarding the possibilities for achieving social processes that exhibit increasing returns.The parsimony and predictability in the study of politics. fuzziness that has marked the use of this concept in social science suggests that the greater range offered by PATH DEPENDENCE AND INCREASING the broader definition has come at a high price in RETURNS analytical clarity.Path dependence has been a victim of what Sartori(1970)called concept stretching.Different Analysts are increasingly inclined to invoke the concept of path dependence,but clear definitions are rare.In types of temporally linked sequences are generated in different ways and have different implications (Abbott practice,usage tends to fluctuate between a broader 1983,1990;Pierson n.d.b).These distinctive kinds of and narrower conception.In the broader version,path social processes,which have been bundled together, dependence refers to the causal relevance of preceding must be disentangled and systematically explored.Lim- stages in a temporal sequence.William Sewell (1996. iting the concept of path dependence to self-reinforc- 262-3),for instance,suggests path dependence means ing processes in no way precludes the investigation of "that what happened at an earlier point in time will other ways in which sequences can matter in explaining affect the possible outcomes of a sequence of events social outcomes.It does encourage clear argument occurring at a later point in time."This usage may about distinct claims.1 entail only the loose and not very helpful assertion that "history matters,"although it may also be presented with more rigor.To take an example offered by An- 1 Another strategy would be to use a broader conception of path dependence and then distinguish subtypes,exploring how these are drew Abbott (1983,131),the contemporary signifi- generated in different ways,with different consequences (Mahoney cance of a group's organization of 40%of its potential n.d.).Because path dependence is often invoked without further membership depends greatly on whether that member- clarification,however,this option seems more problematic.Ulti- ship used to be 100%or 10%.Note that Sewell's mately,such definitional disputes remain intractable.What is critical is that researchers be clear and consistent about what they mean definition involves no necessary suggestion that a par- when they employ the concept and recognize the importance of ticular path is difficult to exit.Rather,the claim is that differentiating among distinct processes. 252
American Political Science Review Vol.94,No.2 However such issues of concept formation are ulti- be possible.We cannot predict ahead of time which mately resolved,there are two compelling reasons for of these possible end-states will be reached. focusing special attention on processes that exhibit 2.Inflexibiliry.The farther into the process we are,the increasing returns.First,such processes characterize harder it becomes to shift from one path to another. many important parts of the social world.Second, In applications to technology,a given subsidy to a social scientists are developing theory that makes the particular technique will be more likely to shift the investigation of the causes and consequences of in- ultimate outcome if it occurs early rather than late. creasing returns a particularly promising area of in- Sufficient movement down a particular path may quiry.These two reasons are developed in later sec- eventually lock in one solution.4 tions of this essay. 3.Nonergodicity.Accidental events early in a sequence The basic logic of increasing returns processes can do not cancel out.They cannot be treated (which is be captured in a simple mathematical illustration.? to say,ignored)as "noise,"because they feed back Imagine a very large urn containing two balls,one into future choices.Small events are remembered. black,one red.Remove one ball,and then return it to 4.Potential path inefficiency.In the long-run,the out- the urn,accompanied by an additional ball of the same come that becomes locked in may generate lower color.Repeat this process until the urn fills up.What pay-offs than a forgone alternative would have. can we say about the eventual distribution of colored balls in the urn?Or about a series of trials in which we To this one can add a general point of particular fill the urn and then start over again one hundred interest to social scientists:These are processes in times? which sequencing is critical.Earlier events matter much more than later ones,and hence different se- In each individual trial we have no idea what the quences may produce different outcomes.In these eventual ratio of red to black balls will be;it could be processes,history matters. 99.9%red,or 0.01%red,or anything in between.If If these characteristics are common in politics,then we were to run 100 trials,we would probably get 100 they carry major implications,both for the kinds of different outcomes. questions we should ask and the kinds of answers we In any particular trial,the ratio will eventually reach should expect to find.Most important,they suggest the an equilibrium.Later draws in a series contribute need to focus on the temporal dimensions of social only minutely to the distribution of balls in the urn. processes.In searching for explanation,we need to Thus,the distribution settles down onto a stable think about causes and effects that are often separated outcome. in time,rather than focus exclusively on synchronic Sequence is thus crucial.Early draws in each trial, explanations (Harsanyi 1960;Stinchcombe 1968).Ex- which have a considerable random element,have a ploring these implications requires a review of recent powerful effect on which of the possible equilibria work on increasing returns in economics. will actually emerge. Mathematicians call this a Polya urn process.Its char- INCREASING RETURNS ARGUMENTS IN acteristic qualities stem from the fact that an element ECONOMICS of chance (or accident)is combined with a decision rule that links current probabilities to the outcomes of Economics has traditionally focused on the search for preceding (partly random)sequences.3 Polya urn pro- unique equilibria.5 The goal is attractive because it cesses exhibit increasing returns or positive feedback. suggests a world of potential predictability and effi- Each step along a particular path produces conse- ciency.Given knowledge of existing factor endowments quences which make that path more attractive for the and preferences,equilibrium analysis will point to a next round.As such effects begin to accumulate,they single optimal outcome.Moreover,because econo- generate a powerful virtuous (or vicious)cycle of mists assume a context of decreasing marginal returns, self-reinforcing activity. this goal is potentially achievable.With decreasing Increasing returns processes have quite intriguing returns,economic actions will engender negative feed- characteristics,which Arthur (1994,112-3)has sum- marized as follows 4 This emerging stability represents a critical distinction between increasing returns processes and chaotic processes,which may gen 1.Unpredictability.Because early events have a large erate no equilibrium.For an interesting discussion of this quite effect and are partly random,many outcomes may different framework,with applications to politics,see Fearon 1996. s Some social scientists,such as Stephen I.Gould,also have been drawn to arguments about path dependence,critical junctures,and 2 The following discussion relics heavily on Arthur 1994,a collection punctuated equilibria in evolutionary biology (see especially krasner of his ground-breaking essays on increasing returns and path depen- 1989;Spruyt 1994).Without denying the relevance of this literature, dence. I find it a less useful point of departure than the economists'focus on 3 This case depicts a specific type of increasing returns process,in increasing returns.Most aspects of politics lack anything like the which the probability of a particular draw precisely equals the ratio mechanism of natural selection that drives Darwinian theory (inter- between the two alternatives in the existing population.Arthur national relations and certain characteristics of electoral systems (1994)shows that many of the features of this case have a greater constitute important exceptions).Furthermore,socially created con- range of application,but not all of them.It is easy to model path structs of norms and formal institutions have no real analog in dependent processes with only two equilibria (e.g.,Hill 1997),which evolutionary theory.These constructs,however,are crucial features probably comes closer to capturing the essence of many path of politics and,as we shall see,are a critical element in social dependent processes in the social world. processes subject to increasing returns. 253
Increasing Returns,Path Dependence,and the Study of Politics June 2000 back,which will lead to a predictable equilibrium.A is essential,as we shall see,because analytically similar sharp rise in oil prices prompts increased conservation, circumstances occur frequently in the world of politics. exploration,and exploitation of other sources of en- Arthur's characteristics provide a foundation for devel- ergy,which will lead to a fall in oil prices.Each step oping hypotheses about when increasing returns pro- away from equilibrium is more difficult than the one cesses are likely to operate in the social world. before.As Arthur (1994,1)summarizes,negative Arthur (1994,112)argues that four features of a "feedback tends to stabilize the economy because any technology and its social context generate increasing major changes will be offset by the very reactions they returns. generate....The equilibrium marks the best'outcome possible under the circumstances:the most efficient use 1.Large set-up or fixed costs.These create a high and allocation of resources." pay-off for further investments in a given technol- During the past fifteen years,however,this decreas- ogy.With large production runs,fixed costs can be ing returns tradition has faced a mounting challenge spread over more output,which will lead to lower Economists have exhibited a growing interest in the unit costs.When set-up or fixed costs are high, idea of increasing returns.On a wide range of sub- individuals and organizations have a strong incen- jects-including the spatial location of production,the tive to identify and stick with a single option. development of international trade,the causes of eco- 2.Leaning effects.Knowledge gained in the operation nomic growth,and the emergence of new technolo- of complex systems also leads to higher returns from gies-path dependence arguments have become prev- continuing use.With repetition,individuals learn alent.The ideas developed in this research are not how to use products more effectively,and their entirely new,but they have been embraced and devel- experiences are likely to spur further innovations in oped by prominent mainstream economists.This work the product or in related activities. has received considerable attention in leading journals. 3.Coordination effects.These occur when the benefits Douglass North,who places great emphasis on such an individual receives from a particular activity arguments in his analysis of the development of mod- increase as others adopt the same option.If tech- ern capitalism,was awarded the Nobel Prize for eco- nologies embody positive network externalities nomics. then a given technology will become more attractive The study of technology has provided the most as more people use it.Coordination effects are fertile ground for arguments based on increasing re- especially significant when a technology has to be turns.As Arthur (1994)and David (1985)have compatible with a linked infrastructure (e.g.,soft- ware with hardware;automobiles with an infrastruc- stressed,under conditions often present in complex. knowledge-intensive sectors,a particular technology ture of roads,repair facilities,and fueling stations) may achieve a decisive advantage over competitors, Increased use of a technology encourages invest- although it is not necessarily the most efficient alterna- ments in the linked infrastructure,which in turn tive in the long run.This occurs because each technol- attracts still more users to the technology. ogy generates higher payoffs for each user as it be- 4.Adaptive expectations.If options that fail to win comes more prevalent.When a new technology is broad acceptance will have drawbacks later on,then subject to increasing returns,being the fastest out of individuals may feel a need to "pick the right horse. the gate (if only for reasons of historical accident) Although the dynamic here is related to coordina becomes critical.With increasing returns,actors have tion effects,it derives from the self-fulfilling charac- strong incentives to focus on a single alternative and to ter of expectations.Projections about future aggre- continue down a specific path once initial steps are gate use patterns lead individuals to adapt their taken in that direction.Once an initial advantage is actions in ways that help make those expectations gained,positive feedback effects may lock in this come true. technology,and competitors are excluded.Path depen- This discussion of technology is important primarily dence arguments have been applied to the develop- because it clarifies a set of relationships characteristic ment of the“QWERTY”typewriter keyboard,.the of many social interactions.New social initiatives- triumph of the light-water nuclear reactor in the such as the creation of organizations or institutions- United States,the battles between Betamax and VHS usually entail considerable start-up costs;individuals, video recorders and between DOS-based and Macin- as well as organizations,learn by doing;the benefits of tosh computers,early automobile designs,and compet- our individual activities or those of an organization are ing standards for electric current.6 often enhanced if they are coordinated or"fit"with the Not all technologies,however,are prone to increas- activities of other actors or organizations;it is fre- ing returns.Arthur (1994)addresses not only the quently important to bet on the right horse,so we characteristics of such processes but also the conditions adapt our actions in light of our expectations about the that give rise to them.Understanding these conditions actions of others. Although arguments about technology are probably the best known,economists have applied similar anal 6 Many of these examples have been contested by critics who deny the empirical claim that superior technologies lost out.Since these yses of increasing returns processes in a striking range criticisms raise broader issues about the usefulness of increasing of economic contexts.Both Krugman (1991)and returns arguments,I will postpone discussion until the end of this Arthur (1994,49-67)point to the role of increasing section. returns in the spatial location of production.Given the 254