Constraining the State:
Introducing Regime Types

Crimes of the State | Week 2, Lecture 3

Professor Julian E. Gerez

April 6, 2026

Roadmap: Constraining the state

  • Last time: the state is a “violence-producing enterprise”
    • All states use coercion
    • The social contract: we give up revenue for security
    • But who protects us from the state?
    • Any questions from last week?
  • Today:
    • What is a regime?
    • What is democracy? What is autocracy?
    • Does regime type change how states behave toward their citizens?

Key concepts

State, government, and regime

  • A state is an entity that uses coercion and the threat of force to rule a given territory
    • The state is a set of institutions
  • A government is the set of people who run the state at any given time
    • Governments come and go; the state persists
  • A regime is the set of rules, norms, and institutions
    • …that determine how the government is constituted and how major decisions are made
    • How power is acquired, exercised, and transferred

The same state can have different governments and different regimes over time

  • Germany: Weimar Republic \(\rightarrow\) Nazi regime \(\rightarrow\) Federal Republic
  • Russia: Tsarism \(\rightarrow\) Soviet Union \(\rightarrow\) Russian Federation

Democracy

What does “democracy” mean?

  • Democracy means “rule by the people” (demos = people, kratos = power)
    • Contrast with “rule by the few”: “aristocracy” (best) or “oligarchy” (few)
    • Contrast with “rule by the one”: “monarchy” (one) or “autocracy” (self)
  • Aristotle classified regimes by who rules and for whose benefit
  • Rule in the common interest is good; rule in the rulers’ self-interest is bad
Number of rulers Good form (for all) Bad form (for rulers)
One Monarchy Tyranny
Few Aristocracy Oligarchy
Many Politeia Democracy
  • For Aristotle, democracy was bad: rule by the many for the many, not the common good
  • Calling something a “democracy” has always been contested
  • How do we define democracy today? Do we know it when we see it?

Procedural vs. substantive definitions

  • Procedural (minimalist): democracy depends on the rules governing the political process
    • Who gets to participate?
    • How are leaders selected?
    • The type of government these procedures produce is irrelevant
  • Substantive: democracy depends on the outcomes it produces
    • Representation, accountability, economic equality, protection of rights

Dahl’s requirements for democracy

Robert Dahl, American political scientist

Robert Dahl (1971)

  1. Freedom to form and join organizations
  2. Freedom of expression
  3. The right to vote
  4. Open eligibility for public office
  5. The right of political leaders to compete for support
  6. Alternative sources of information
  7. Free and fair elections
  8. Policymaking institutions depend on votes/expressed preferences

Are these procedural or substantive requirements?

Which requirement could you most justify removing?

Dahl’s two dimensions of democratization

  1. Inclusiveness (participation)
    • The share of the population who involved in the political process
  2. Liberalization (public contestation)
    • The degree to which state policies are contested publicly
Liberalization (public contestation) Inclusiveness (participation) Inclusive hegemony Polyarchy Closed hegemony Competitive oligarchy

Inclusive hegemony: North Korea

  • High inclusiveness (participation)
    • Universal suffrage for all citizens >17
    • The 2026 election: 99.99% turnout
    • Official explanation for the missing 0.01%: some citizens were “abroad, or working at sea”
  • Low liberalization (public contestation)
    • Single slate of candidates, no opposition

North Korean elections

March 17, 2026

687 Workers' Party of Korea

Party % Seats
Workers’ Party & allies 99.93 687
Against 0.07 —
Turnout 99.99

Competitive oligarchy: Apartheid South Africa

  • High liberalization (public contestation)
    • Competitive elections with genuine alternation in power
    • Multiple parties, real opposition
  • Low inclusiveness (participation)
    • Non-white citizens barred from voting
    • Elections were real, but only for a fraction of the population

South African government in the 1940s

Other potential defining characteristics of democracy

  • Rule of law: the law is the supreme authority, not individuals

    • Democracy is a system where decisions made through democratic processes are binding
    • Equality under the letter of the law

  • Alternation of power: “democracy is a system where people lose elections”

    • The democratic bargain: when I win I can change policy; when I lose I accept that others will

    • Democracy falters when those in power refuse to leave or losers refuse to accept results

Dennis Hastert

Patricio Aylwin and Augusto Pinochet

Democratic backsliding

  • Democracy is not just “elections,” it requires a bundle of institutions
  • Any one of Dahl’s conditions might be degraded without formally abolishing democracy
    • Gerrymander districts \(\rightarrow\) undermines expressions of preference
    • Restrict press freedom \(\rightarrow\) undermines alternative sources of information
    • Target opposition with criminal charges \(\rightarrow\) undermines the right of politicians to compete
  • Democratic backsliding: the gradual erosion of democratic institutions from within

Fidesz performance in Hungarian parliamentary elections

Election year Share of eligible voters Share of votes cast Share of seats won
2010 41.5% 53.1% 68.1%
2014 26.6% 44.9% 66.8%

Elections will be held in Hungary on April 12, 2026

Classifying autocracies

  • Monarchical
    • Rely on kin and family network to come to power and stay in power
  • Military
    • The head of government is a current or former member of the armed forces
    • May be ruled by committee or junta
  • Civilian (a residual category)
    • Includes single-party regimes (e.g., China) and electoral autocracies (e.g., Russia)

King Mswati III of Eswatini in traditional ceremonial dress

Mswati III (Eswatini)
Chosen king by his father’s wives at age 14

Min Aung Hlaing, commander of the Myanmar military, in full military uniform adorned with medals

Min Aung Hlaing (Myanmar)
Invaded capital city; arrested civilian leaders

Xi Jinping, General Secretary of the Chinese Communist Party, in a suit and tie

Xi Jinping (China)
Appointed leader of Communist Party after 30+ year political career

Does regime type matter?

The puzzle

  • We now know what democracies and autocracies are
  • But does it actually matter for citizens which type of regime they live under?
  • Intuitive answer: yes, democracies are better
  • But why? Mancur Olson (1993) offers an answer

Korean peninsula from space

Mancur Olson

Social order is a public good

Excludable (can prevent non-payers) Non-excludable (cannot prevent non-payers) Rivalrous (use reduces supply) Non-rivalrous (use doesn't reduce supply) Private good Excludable + rivalrous A sandwich, a car, a seat on a train Common-pool resource Non-excludable + rivalrous Forests, fisheries, a public road Club good Excludable + non-rivalrous Netflix, a toll road, a private park Public good Non-excludable + non-rivalrous Clean air, a lighthouse, democracy, social order
  • A public good is non-excludable and non-rivalrous
    • You cannot stop people from benefiting once it exists
    • One person’s enjoyment does not reduce another’s
  • Creates a free-rider problem: each individual is incentivized to let others pay for the good
  • The more people in the group, the worse the problem

Collective action in small vs. large groups

  • In small groups, voluntary cooperation can be feasible
    • Each person receives a significant share of the benefits
    • Each person’s behavior is visible to others: defectors can be identified and sanctioned
    • Example: a hunter-gatherer band maintaining order by consensus
    • Example: doing the dishes with one other roommate, your policy brief
  • In large groups, voluntary cooperation breaks down
    • One person in a million receives one-millionth of the benefit of a public good
    • But bears the full cost of contributing to it
    • Easier to hide individual defection
    • Harder to punish defectors
    • So the rational individual free-rides (enjoys the good without contributing)
    • If everyone reasons this way, the good never gets produced

So where does order come from?

Olson’s thought experiment

  • Imagine a society with no state: only roving bandits
  • Roving bandits plunder indiscriminately
    • If you produce anything, it will be stolen
    • So no one produces anything …and eventually there is nothing left to steal
  • This is a terrible equilibrium for everyone, including the bandits
  • Even a self-interested bandit can do better

Pirates plundering a Spanish colonial town

Feng Yuxiang, Chinese warlord

The stationary bandit

  • A rational bandit realizes: settle down
    • Monopolize theft in one territory
    • But what happens if you fully plunder? Nothing to take in the future
    • Take theft in the form of regular taxation rather than plunder
    • Protect subjects from other bandits
  • Now subjects have an incentive to produce: they keep what remains after taxes
  • The bandit collects more in taxes than he ever could through plunder
  • A warlord becomes a king

“The rational, self-interested leader of a band of roving bandits is led, as though by an invisible hand, to settle down, wear a crown, and replace anarchy with government.” — Olson (1993)

The encompassing interest

  • The stationary bandit now has an encompassing interest in their territory
    • They receive a share of everything produced there
    • So they benefit when their subjects are productive
    • This creates an incentive to provide public goods: roads, courts, security
    • Public goods increase productivity, benefiting the bandit and the people

This is the logic of the “village monarchist”:

“Monarchy is the best kind of government because the King is then the owner of the country. Like the owner of a house, when the wiring is wrong, he fixes it.”

There is some truth to this, but the analogy breaks down…

The grasping hand

  • The autocrat is not like the owner of a single house
  • They are the owner of all wealth in the country
  • They have an incentive to charge monopoly rents
    • Like a landlord who owns every apartment in the city
    • They can charge more than in a competitive market
    • …because you have nowhere else to go
  • But even the autocrat has a limit:
    • Squeeze too hard and subjects stop producing
    • So they stop just short of killing the goose
    • This is the revenue-maximizing rate
  • Subjects are better off than under anarchy
  • …but the autocrat extracts maximum possible surplus

The palace of Versailles

The ornate exterior of the palace of Versailles, depicting a large building, huge fountain, and statues The ornate interior of the palace of Versailles, depicting the Hall of Mirrors including gold, ceiling paintings, and chandeliers

Why democracy extracts less

  • A democratic majority, like the autocrat, controls tax collection

  • But unlike the autocrat, the majority also earns income in the market

  • So when taxes are too high, the majority loses twice:

    • Lower tax revenue
    • Lower market income

  • A rational democratic majority will therefore always extract less than a rational autocrat

  • This is not just about taxes, it is about institutions

  • The institutions needed for lasting democracy…

    • Rule of law
    • Independent courts
    • Property rights, etc.

  • …are exactly the institutions that constrain state predation

  • Democracy and constraints on state power come as a package

Think-pair-share: stationary and roving bandits

  • What does Olson imply about the relationship between the state and its citizens?
  • Does Olson’s logic change how you think about what “crimes of the state” means?
  • Olson’s argument is about development… does the same logic apply to state violence?
  • How is Olson’s argument different from Tilly’s?

Mobutu Sese Seko

Reflections: regime type and state crimes

  • Regime type is not just an abstract concept
  • It shapes the structural incentives rulers face
    • Autocrats are structurally incentivized to over-extract
    • Democratic majorities are structurally less incentivized to do so
  • But this is not the whole story:
    • Democracies also commit crimes of the state
    • Some autocrats restrain themselves; some democracies do not
    • Hybrid regimes blur the picture further

Regime type is a starting point, not a full explanation

Who really constrains the state?

Fragility of constraints on state power

  • Possible constraints on state power:

    • Elections
    • Courts
    • Media
    • Citizens / protest
    • International actors
    • Economic elites
  • Which of these is easiest for a regime to undermine?
  • Would your answer change in an autocracy vs. a democracy?

When domestic checks fail, human rights try to hold the state accountable

For Wednesday: human rights

  • We have seen how regime type shapes incentives
  • But incentives may not be enough: states also need to be constrained by law
  • Wednesday: the international human rights framework as a mechanism of constraint
  • Questions?

Read before Wednesday:

United Nations General Assembly. 1948. Universal Declaration of Human Rights. Resolution 217A. December 10.

and

Langlois, Anthony. 2016. “Normative and Theoretical Foundations of Human Rights.” In Human Rights: Politics and Practice, ed. Michael E. Goodhart, 11–26. Oxford University Press.

(Both are available on Perusall via Canvas)

Crimes of the State | Spring 2026