Institution
An institution is one of the most common things in human life, yet one of the least examined. Handshakes, marriage, money, prisons, languages, the rule of the road: all of these fall under the same broad concept that social scientists have spent decades trying to pin down. What is an institution, exactly? How do institutions come into being, and what makes some survive for centuries while others collapse within a generation? And if institutions are supposed to serve us, why do they so often seem impossible to change even when everyone agrees they are broken?
Sociologist Emile Durkheim described sociology itself as the "science of institutions, their genesis and their functioning." That framing suggests something important: institutions are not background noise. They are the central object of inquiry for political science, anthropology, economics, and sociology alike. Historians study their founding, growth, decay, and development as part of political, economic, and cultural history. What unfolds in the chapters ahead is that story: how institutions form, why they persist, what happens when they weaken, and what it costs a society to be stuck with rules it cannot escape.
Wolfgang Streeck and Kathleen Thelen described institutions as "building blocks of social order": collectively enforced expectations about how specific categories of actors should behave. That phrasing matters. An institution is not just a rule; it is a rule that society enforces together, and one that creates mutually related rights and obligations.
Douglass North offered a tighter formulation: institutions are "humanly devised constraints that shape interaction." North's framing places the emphasis on constraint. Institutions do not merely describe what people do; they narrow what people can do and what they expect of each other.
Jack Knight's Rational Choice definition adds another layer: a valid institution requires that knowledge of its rules be "shared by the members of the relevant community or society." A rule that only one person knows is not an institution. Shared knowledge is the condition.
Robert Keohane pushed the definition toward durability, calling institutions "persistent and connected sets of rules (formal or informal) that prescribe behavioral roles, constrain activity, and shape expectations." Samuel P. Huntington compressed the same idea even further: institutions are "stable, valued, recurring patterns of behavior."
Not everyone agrees on where to draw the boundary. The most expansive definitions count informal practices like handshakes. The narrowest demand written laws, specified enforcement mechanisms, and complex organizational structures. Geoffrey M. Hodgson argued that an institution should not be confused with behavior itself; institutions are "integrated systems of rules that structure social interactions," not the behavior those rules produce. That distinction matters for understanding how institutions change, and how they fail.
Formal institutions are the ones people can point to: constitutions, legal codes, organizational charters. Informal institutions are the ones people absorb without reading. They are socially shared rules, often unwritten, known by nearly everyone in a given country. Informal practices are frequently described as cultural, and clientelism or corruption are sometimes labeled as part of a political culture, but scholars caution against treating informal institutions as merely cultural artifacts. They shape behavior the same way formal institutions do.
The relationship between the two is often closely intertwined. Informal institutions frequently step in to prop up inefficient formal ones. But because informal institutions lack a center that coordinates their actions, changing them is a slow and lengthy process.
Economist Paul A. David introduced the concept of institutional lock-in through a 1985 article titled "Clio and the Economics of QWERTY." His subject was technology: how a specific technology can dominate a market even when it is not the most efficient option available, simply because early adopters forced later actors to conform, regardless of their natural preferences. W. Brian Arthur extended that analysis to institutions. Just as technologies lock in, institutions encoded in law, policy, or social regulation can become embedded so deeply in social and economic frameworks that changing them grows progressively harder over time.
Arthur noted that institutional lock-in, though often predictable in retrospect, is difficult to undo. The costs of changing are not just political. They are structural. The very organizations that benefited from the original arrangement have every incentive to preserve it, and their continued adaptation to the existing framework only tightens the hold the institution has on the surrounding society.
Daron Acemoglu identified three characteristics that distinguish good institutions: enforcement of property rights, constraints on the actions of elites, and equal opportunity for broad segments of society. That definition aligns closely with the World Bank's concept of "market-supporting institutions," which the bank's Worldwide Governance Indicators, introduced in 1999, were designed to measure. The bank's seminal report on governance concluded that "there is a strong causal relationship from good governance to better development outcomes."
Steven Levitsky and Maria Victoria Murillo approached the same question from the opposite direction, defining weak institutions as those whose "rules are frequently violated or changed." They cited labor regulation in Latin America and party systems in Argentina and Peru as examples. Consistent with that assessment, Latin American countries' scores on the Worldwide Governance Indicators fell below the global median, while top-ranking countries included Denmark and the UK.
Levitsky and Murillo argued that institutional strength depends on two factors: stability and enforcement. When rules are unstable or go unenforced, actors cannot depend on each other to behave predictably, which erodes the foundation for collective action and collaboration. Weak enforcement creates what they describe as a slippery slope: once some laws are routinely ignored, the erosion tends to spread.
Lars Feld and Stefan Voigt found empirical support for a counterintuitive result: real GDP growth per capita correlates positively with de facto judicially independent institutions, not de jure ones. The formal guarantee of independence, in other words, matters less than whether independence is actually practiced. Levitsky and Murillo used a pointed metaphor for institutions created mainly to satisfy international expectations without genuine enforcement: window dressing. Such arrangements give power holders the appearance of compliance while preserving their practical freedom to ignore the rules entirely.
Douglass North's 1990 book, "Institutions: Institutional Change and Economic Performance," laid out two basic paths: institutions can be deliberately created, as a country's constitution might be, or they can evolve gradually as societies change. The detection of that second kind of evolution is notoriously difficult, North noted, because societal changes move slowly even when institutional change appears rapid.
Levitsky and Murillo examined the role of timing. When actors create institutions under pressure or within a compressed timeframe, they have less opportunity to calculate the downstream effects of the rules they are writing, which makes the resulting institutions more vulnerable to weakness from the start.
Christopher Kingston and Gonzalo Caballero focused on the role of gradual societal change in determining which new institutions take hold. The choices individual actors make within a society influence which rules gain enough support to survive, spread, and become self-sustaining. People's interests are not passive in this process: they actively foster groups that benefit from the existing rules, reinforcing a status quo and making further change harder.
Pavlovic identified three conditions under which institutions can take root even in poor societies with no democratic background: if electoral institutions guarantee multiple elections that are widely accepted, if military power is held in even equilibrium, and if institutional arrangements allow different actors to take power. Without at least one of these conditions, the rules may exist on paper without generating the compliance that gives them real force.
Yuen Yuen Ang proposed a coevolutionary model that challenged both the "good institutions first" and "growth first" schools of thought. Her alternative describes a three-step sequence: build markets using normatively weak institutions; upgrade those institutions; then use modern institutions to preserve markets. She applied this framework to post-1980s China and 19th-century United States, arguing that neither fits the standard models cleanly.
North described path dependence as the mechanism by which institutional patterns endure long past the circumstances that created them. Choices made at "critical junctures," moments he compared to forks in the road, narrow the range of possible futures. Once a path is taken, returning to the original decision point becomes progressively more difficult.
James Mahoney tested this theory in the context of national regime change in Central America. He found that liberal policy choices made by Central American leaders in the 19th century constituted the critical juncture that produced the divergent development levels visible today. The specific choices made during liberal reform created self-reinforcing institutions whose effects compounded over time.
North, Wallis, and Weingast divided societies into two social orders: open access orders, which roughly a dozen developed countries fall into today, and limited access orders, which account for the rest. Transition between these orders does not happen simply by transplanting open-access institutions into new contexts. Those institutions tend to be coopted by powerful elites for self-enrichment. Real transition happens only when it is in the interest of the dominant coalition to widen access.
Acemoglu, Johnson, and Robinson built a framework for understanding how this cycle perpetuates itself. De jure and de facto political power are determined by preexisting institutions and the distribution of resources. Those two forms of power shape economic institutions in the present period and political institutions in the next. Economic institutions in turn determine how resources are distributed in the following period, and the cycle repeats. Institutional change, under this model, is endogenous: the system tends to reproduce conditions that favor those already in control of it.
Ian Lustick proposed borrowing the concept of natural selection from biology to explain why institutions often persist in harmful forms even when leaders and members alike recognize the problem. Viewing institutions as existing within a fitness landscape, Lustick argued that gradual improvement is analogous to hill-climbing. Eventually, institutions can become stuck on local maxima, positions where further improvement requires a temporary decline in fitness, meaning adopting policies that cause short-term harm before longer-term gains become visible.
Lustick cited Amyx's analysis of Japan's economic trajectory as his primary example. During the so-called Lost Decade, Japanese experts were not unaware of what was causing the country's economic decline. The problem was that correcting course would have required policies that first caused short-term harm to the Japanese people and government. Japan had climbed to a local maxima set by the economic landscape of the 1970s and 1980s, and without institutional flexibility, it could not adapt to conditions that had since changed.
Lustick extended the lesson to the United States, observing that politicians who run for elected office face strong incentives to prioritize policies with visible short-term results. The mismatch between short-term benefits and the long-term institutional adaptations needed for durable change is not accidental; it is built into the selection pressures operating on political actors.
Critics pushed back on several fronts. David Sloan Wilson argued that Lustick needed to distinguish more carefully between multilevel selection theory and evolution on multi-peaked landscapes. Bradley Thayer raised a harder objection: the fitness landscape metaphor only holds if one institution can be judged objectively better than another. For economic prosperity, that measure is relatively tractable. For the quality of freedom in a society, or the subjective wellbeing of its members, objective comparison becomes much harder to establish.
The word "institutionalization" carries two very different registers. In social theory, it refers to the process of embedding a concept, a social role, a value, or a mode of behavior within an organization or a society as a whole. In everyday speech, the same word often refers to confining a vulnerable individual to a mental institution, a usage that carries connotations of coercion, damage, and the crushing weight of inflexible systems.
Both uses point to the same underlying tension. Institutions are systems of rules that structure social life. When those rules are enforced fairly and consistently, they generate the trust that makes collective action possible. When they are enforced coercively, selectively, or not at all, they erode that trust, sometimes in ways that are difficult to recover from.
The sources of institutional distrust are specific. Institutional abuse, institutional discrimination, bureaucratic drift, and politicization are each identified as factors that reduce trust in institutions. That erosion is not abstract. Acemoglu, Johnson, and Robinson documented cases in which weak enforcement led to the deterioration of democratic institutions in Madagascar and the erosion of economic structures in China. For Lacatus, writing on national human rights institutions in Europe, membership in GANHRI correlates with institutions becoming stronger over time, showing a general pattern of isomorphism toward stronger safeguards for durability. That finding points toward one concrete mechanism through which institutional trust, once lost in one domain, can be deliberately rebuilt through sustained, externally validated compliance.
Common questions
What is an institution in social science?
An institution is a humanly devised structure of rules and norms that shape and constrain social behavior. Definitions vary by discipline, but all generally require some level of persistence and continuity, and examples range from laws and formal organizations to informal social conventions and norms.
Who defined institutions as humanly devised constraints that shape interaction?
Douglass North defined institutions as "humanly devised constraints that shape interaction." North argued they are critical determinants of economic performance, affecting the costs of exchange and production, and that small historical and cultural features can drastically change the nature of a given institution.
What is institutional lock-in and who developed the concept?
Institutional lock-in describes the process by which a rule, law, or social arrangement becomes so embedded in social and economic frameworks that it is very difficult to change, even when alternatives exist. Economist Paul A. David introduced the concept in a 1985 article titled "Clio and the Economics of QWERTY," and W. Brian Arthur extended it from technology to institutions.
What are the three characteristics of good institutions according to Daron Acemoglu?
According to Daron Acemoglu, good institutions have three key characteristics: enforcement of property rights, constraints on the actions of elites, and equal opportunity for broad segments of society. The World Bank's Worldwide Governance Indicators, introduced in 1999, were designed to measure these dimensions.
What is path dependence in institutional theory?
Path dependence is the idea that choices made at critical junctures in an institution's history narrow the range of possible future outcomes, making it progressively harder to return to the original decision point. James Mahoney studied this in the context of Central America, finding that liberal policy choices by 19th-century leaders produced the divergent development levels seen in those countries today.
How did Ian Lustick apply natural selection to institutional change?
Ian Lustick argued that institutions exist within a fitness landscape and that gradual improvement is analogous to hill-climbing. Institutions can become stuck on local maxima, where further improvement requires short-term harm before long-term gains emerge. He used Amyx's analysis of Japan's Lost Decade as his primary example, where experts knew the needed fixes but could not enact them without imposing short-term costs.
All sources
84 references cited across the entry
- 1journalInstitutionsDouglass C. North — 1991
- 2citationIntroduction: Theories of InstitutionsCambridge University Press — 2022
- 3webSocial InstitutionsMetaphysics Research Lab, Stanford University — 2014
- 6citationDefinitions of InstitutionsCambridge University Press — 2022
- 7journalEconomic Institutions and the State: Insights from Economic HistoryHenning Hillmann — 2013
- 8journalRational Actors, Equilibrium and Social InstitutionsRandall Calvert — 1995
- 9bookBeyond Continuity: Institutional Change in Advanced Political EconomiesWolfgang Streeck et al. — Oxford University Press — 2005
- 10bookInstitutions and social conflictJack Knight — Cambridge University Press — 1992
- 11bookInstitutions, Institutional Change, and Economic PerformanceDouglass North — Cambridge University Press — 1990
- 12journalInstitutions as a Fundamental Cause of Long-Run GrowthDaron Acemoglu et al. — 2005
- 13journalInstitutions, Human Capital, and DevelopmentDaron Acemoglu et al. — 2014
- 14journalVariation in institutional strengthSteven Levitsky et al. — 2009
- 15citationPaths to Inclusive Political InstitutionsDaron Acemoğlu — Springer — 2016
- 16journalClio and the Economics of QWERTYPaul A. David — May 1985
- 17journalCompeting technologies, increasing returns, and lock-in by historical eventsW. Brian Arthur — 1989
- 18journalInternational Institutions: Two ApproachesRobert O. Keohane — 1988
- 19bookPolitical Order in Changing SocietiesSamuel P. Huntington — Yale University Press — 1996
- 20journalA Theory of Endogenous Institutional ChangeAvner Greif et al. — 2004
- 21bookExplaining Institutional Change: Ambiguity, Agency, and PowerCambridge University Press — 2009
- 22journalInformal Institutions and Comparative Politics: A Research AgendaGretchen Helmke et al. — 2004
- 23journalRoot Causes: A historical approach to assessing the role of institutions in economic developmentDaron Acemoglu — June 2003
- 24bookWorld Development Report 2002: Building Institutions for MarketsWorld Bank — World Bank — 2002
- 25webGovernance mattersKaufmann, Daniel,Kraay, Aart,Zoido-Lobaton Pabl
- 26webCorruption MattersDaniel Kaufman — 2015
- 28journalSocial learning and economic behavior: The process of economic socializationR. B. Denhart et al. — 1971
- 29journalBroad and narrow socialization: The family in the context of a cultural theoryJ. J. Arnett — 1995
- 30journalMeasuring Electoral Democracy with V-Dem Data: Introducing a New Polyarchy IndexJan Teorell et al. — Elsevier BV — 2016
- 31journalAlienation in prison: An examination of the work relation of prison guardsE. D. Poole et al. — 1981
- 32journalThe role of social energy in prisonA Carmi — 1983
- 34journalLanguage, culture, gender, and academic socializationN Morita — 2009
- 35journalWhere is the child's environment? A group socialization theory of developmentJ. R. Harris — 1995
- 36journalMedia as Social InstitutionArt Silverblatt — 2004
- 37journalGender As Social InstitutionPatricia Martin — 2004
- 38journalVideo Games as Social InstitutionsDragos Obreja — 2023
- 39bookDesigning Institutions for Environmental and Resource ManagementE. Elgar — 1998
- 40journalIncreasing Returns, Path Dependence, and the Study of PoliticsPaul Pierson — 2000-01-01
- 41bookGender and Rural DevelopmentOlanike F Deji — LIT Verlag — 2011
- 42journalThe 'credibility thesis' and its application to property rights: (In)Secure land tenure, conflict and social welfare in ChinaPeter Ho — September 2014
- 43journalThe political economy of 'policy credibility': the new-classical macroeconomics and the remaking of emerging economiesIlene Grabel — 2000
- 44journalVariation in Institutional StrengthSteven Levitsky et al. — 2009
- 45bookInstitutions, Institutional Change and Economic PerformanceDouglass North — Cambridge University Press — 1990
- 46journalParchment, Equilibria, and InstitutionsJohn M. Carey — 2000
- 47bookThe Fate of Africa's Democratic Experiments: Elites and InstitutionsRichard R. Marcus — Indiana University Press — 2005
- 48journalAdaptive Informal Institutions and Endogenous Institutional Change in ChinaKellee Sing Tsai — 2006
- 49journalDe Facto Versus de Jure Political Institutions in the Long-Run: A Multivariate Analysis, 1820-2000Peter Foldvari — 2017
- 50journalEconomic Growth and Judicial Independence: Cross-Country Evidence Using a New Set of IndicatorsLars P. Feld et al. — 2003
- 51journalInformal Institutions and Comparative Politics: A Research AgendaGretchen Helmke et al. — 2004
- 52bookInstitutions, Institutional Change and Economic PerformanceDouglass C. North — Cambridge University Press — 1990
- 53bookInstitutions, Institutional Change and Economic PerformanceDouglass C. North — Cambridge University Press — 26 October 1990
- 54journalComparing theories of institutional changeChristopher Kingston et al. — August 2009
- 55journalSpontaneous OrderRobert Sugden — 1989
- 56journalHow Democratic Institutions EmergeDušan Pavlović — 2010
- 57bookSocial InstitutionsMichael Hechter — Routledge — 1990
- 58journalThe Rational Design of International InstitutionsBarbara Koremenos et al. — 2001
- 59journalInstitutionalized Organizations: Formal Structure as Myth and CeremonyJohn W. Meyer et al. — 1977
- 60journalThe Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational FieldsPaul J. DiMaggio et al. — 1983
- 61bookHow China Escaped the Poverty TrapYuen Yuen Ang — Cornell University Press — 2016
- 62journalAdaptive Political Economy: Toward a New ParadigmYuen Yuen Ang — 2025
- 64journalAn introduction to institutions and institutional changeDouglass C. North — June 5, 2012
- 65journalExplaining Institutional Change: Policy Areas, Outside Options, and the Bretton Woods InstitutionsPhillip Lipscy — 2015
- 66journalCompeting Technologies, Increasing Returns, and Lock-In by Historical EventsW. Brian Arthur — 1989
- 67journalDemocratic Transitions, Institutional Strength, and WarEdward Mansfield et al. — 2002
- 68bookThe Soldier and the State: the theory and politics of civil-military relationsSamuel P. Huntington — Harvard University Press
- 69citationVariation in Institutional StrengthLevitsky et al. — 2009
- 70journalExplanations of Institutional Change: Reflecting on a "Missing Diagonal"Johannes Gerschewski — 2020
- 71journalMaking Sense of the Design of International InstitutionsErik Voeten — 2019-05-11
- 72bookHow China Escaped the Poverty TrapYuen Yuen Ang — Cornell University Press — 2016
- 73webBook Review: How China Escaped the Poverty Trap by Yuen Yuen Ang - LSE Review of BooksRose Deller — 2017-10-16
- 75webHow Development Actually Happened: An Excerpt from How China Escaped the Poverty TrapYuen Yuen Ang — 2026
- 76bookInstitutions, Institutional Change and Economic PerformanceDouglass North — Cambridge University Press; 59262nd edition — October 26, 1990
- 77journalVariation in Institutional StrengthSteven, Maria Levitsky, Murrillo — 2009
- 79webEU bans TikTok from official devices across all three government institutionsXiaofei, Eve, James Xu, Brennan, Frater — CNN — March 2023
- 80webSilvergate Closes SEN Platform Institutions Used to Move Money to Crypto ExchangesIan, Nikhilesh Allison, De — Yahoo
- 81journalDemocracy, Governance, and Economic Growth: Theory and EvidenceFrancisco L. Rivera-Batiz — 2002
- 83journalInstitutional Rigidity and Evolutionary Theory: Trapped on a Local MaximumIan Lustick — 2011
- 84bookJapan's Financial Crisis: Institutional Rigidity and Reluctant ChangeJennifer Amyx — Princeton University Press — 2004
- 85journalFifty Years of Declining Confidence & Increasing Polarization in Trust in American InstitutionsHenry E. Brady et al. — American Academy of Arts & Sciences, The MIT Press — 2022