The Politics of Corruption

Lecture 1: Welcome, and What Is Corruption?

Bogdan G. Popescu

Tecnológico de Monterrey

The Course

What you will be able to do

  • Distinguish bribery, fraud, clientelism, patronage, state capture
  • Explain why corruption persists in some countries
  • Weigh what the evidence says corruption costs
  • Compare how some countries reduced corruption, and why
  • Read a study and say what it actually established

The path

  • Part I — concepts and evidence (Sessions 1–5)
  • Midterm review S6 · MidtermThu, Sep 24 · debrief S7
  • Part II — how corruption grew and receded (Sessions 8–10; S8 before the midterm)
  • Part III — corruption in a globalized world (Sessions 11–13)
  • Final review S14–S15 · FinalThu, Nov 26
  • No class during Semana Tec: Sep 14–18 and Oct 26–30

Logistics

  • Meetings: Monday lecture · Thursday discussion and presentations
  • Term: Aug 10 – Nov 26, 2026 · no class Semana Tec (Sep 14–18, Oct 26–30)
  • Office Hours: by appointment — bgpopescu@tec.mx
  • Materials: course site, one page per session — slides and readings (PDF)
  • Readings: Required = done before class, examinable · Recommended = depth. Both are used for presentations
  • Exams: midterm Thu, Sep 24 · final Thu, Nov 26 — in class, closed book

Grading and rules

Component Weight
Contributions 33%
Midterm exam 33%
Final exam 33%

Contributions — the average of:

  • Presentation — one 10-minute talk on an assigned reading
  • Presence — attendance across the term
  • Participation — what you contribute to discussion
  • Questions — two per session, before 11pm Wednesday
  • Quality of questions — judged on substance, not submission

Questions Every Session

Submit two questions on the session’s readings before 11pm Wednesday, ahead of the Thursday discussion. They guide it.

Each question should be between 50 and 100 words.

Begin by referencing a specific part of the text or its overarching claim. Then ask a critical question that opens discussion rather than testing recall.

Questions are graded on quality, not just submission.

Example Question

Note

Nye argues that corruption can sometimes help a poor country develop, because bribes let people get around slow official rules and finally get things done. But if bribes are what make slow rules profitable, why would any official ever simplify them? Someone who earns from delay has a reason to create more delay. Nye treats red tape as a fixed obstacle that corruption helps people escape. What if corruption is what produces the red tape in the first place? How would we tell those two apart?

Why is it good?

  • 50–100 words
  • Points at one specific claim, not “the reading” in general
  • Asks a question instead of summarizing
  • Offers a rival explanation — corruption may cause the red tape, not cure it

Presentations

  • One per student: an assigned reading, on a Thursday — starting Aug 20, two or three of you each week
  • Format: 10 minutes
  • Structure: clear introduction, main points, conclusion
  • Critical analysis: go beyond summary — what the author does not address, the limits of the evidence, rival explanations
  • Close on a verdict: well supported · plausible but not established · not established by these data
  • Visual aids: legible text, relevant visuals, no overloaded slides
  • Preparation: practice for time; do not read verbatim from slides or notes
  • Engage: be ready for 2–3 questions from peers or the instructor

How presentations are graded

Criterion Points
Critical analysis and verdict — the unstated premise, rival explanations, a defensible verdict 30
Content knowledge — the claim that matters, and the evidence behind it 20
Selection and focus — one line of argument, developed, within the time 15
Organization — introduction, main points, conclusion 15
Clarity of expression 10
Engagement and response — speaking to the room, handling questions 10

Analysis outweighs coverage. At ten minutes, choosing well is the skill.

Today’s game: “Is this corruption?”

Four real situations — three from Latin America, one from the United States.

For each one, you will vote:

  • Clearly corrupt
  • Not corrupt
  • It’s complicated

Write your four votes down — we will return to them at the end.

Case 1 — A police stop on the street

Mexico

  • A driver runs a red light in the city
  • The official fine: money, points, a lost morning
  • The officer hints: “Podemos arreglarlo aquí”
  • Two hundred pesos change hands; no ticket, no record
  • Both sides walk away better off — today

Corrupt? Who exactly is the victim?

A Mexico City transit police officer signals a car to stop on Insurgentes Avenue.

Transit officer stopping a car, Mexico City

Photo: ProtoplasmaKid, Wikimedia Commons, CC BY-SA 4.0.

Case 2 — Despensas before the vote

Mexico

  • Election week, a low-income neighborhood
  • Party workers hand out grocery packages and store gift cards
  • Names are noted; benefits, it is hinted, depend on winning
  • 2012: the “Soriana card” accusations — never proven in court
  • Nobody was coerced; campaigns give things away everywhere

Corrupt? A gift, a wage, or a purchase?

Case 3 — The revolving door

United States

  • A senior financial regulator supervises a major bank for years
  • Soon after they leave office, that bank hires them — at triple pay
  • Entirely legal after a short cooling-off period
  • Expertise should be employable; who else knows the rules so well?
  • But: did the prospect of the job soften supervision earlier?

Corrupt? Nothing was ever exchanged — perhaps.

Case 4 — La Estafa Maestra

Mexico

  • 2013–2014: eleven Mexican government agencies awarded contracts to eight public universities
  • The universities were supposed to carry out public projects
    • instead they subcontracted most of the work to private companies
    • Of the 186 companies hired, 128 were shell companies or irregular firms, or could not be located
  • Journalists traced 7,670 million pesos of diverted public money

Corrupt? Here, almost nobody hesitates. Why?

Source: Animal Político & Mexicanos Contra la Corrupción y la Impunidad (2017).

Exercise 1 — Vote and defend

≈ 5 minutes

Prompt. In pairs: rank the four cases from most clearly corrupt to least clearly corrupt.

  1. Compare your votes on the four cases.
  2. Why is it corrupt, or why not?

Keep your test visible — we are about to stress-test it.

The four cases, side by side

Case Where In one line
1 A police stop on the street Mexico red light · 200 pesos · no ticket, no record
2 Despensas before the vote Mexico election week · groceries and gift cards · names noted
3 The revolving door United States regulator supervises a bank · the bank then hires them
4 La Estafa Maestra Mexico contracts via public universities · shell companies · 7,670 million pesos

Rank these. Keep your test in one sentence.

What did our disagreement reveal?

Call out your pair’s test — let’s collect them.

Tests that usually surface, and where they clash:

  • Legality — some cases were illegal but small, others legal but vast
  • Harm — diffuse in some cases, brutal in others
  • Motive — some actors believed they were doing their jobs

So what? Intuition alone cannot draw the line. There is no neutral way to count corruption — before we can measure it, prosecute it, or reform it, we have to decide what counts.

Part II: Defining Corruption

How three institutions approach corruption

Institution How it frames corruption Where it draws the line
World Bank (1997) “the abuse of public office for private gain” the public sector
Transparency International “the abuse of entrusted power for private gain” public and private power
OECD (2008) no single definition; specifies criminal acts — bribery of officials, trading in influence prosecutable offences
  • All three come from reform institutions — built for measuring and prosecuting, not for settling what corruption is

Sources: World Bank (1997); Transparency International (2025); OECD (2008).

Stress test — the revolving door

Back to Case 3: the regulator supervises a bank, then the bank hires them.

Did corruption occur?

  • World Bank — only if they abused public office while still in government
  • Transparency International — only if they abused the power entrusted to them
  • OECD — taking the job is not, in itself, a bribery offence

Same facts. Not one definition gives a clean yes or no.

Three questions to ask about any case

When you think something might be corrupt, ask:

Three questions in sequence: who had power, how was it misused, and who benefited.
  • Case 1, the traffic stop: police authority → the ticket is never written → 200 pesos to the officer
  • The technical names for the same three: entrusted powerdeviation from dutyprivate gain
  • Most definitions wrestle with the same three problems — but disagree about what counts as power, misuse, and gain
  • The hardest is the middle one: what standard tells us power was misused?

Three ways to define corruption

Three scholarly answers to that middle question — misused by what standard?

Approach Conduct is corrupt when it violates… Example
Public office the formal duties of a public official Nye (1967)
Public interest the broader interests of society Friedrich
Public opinion what citizens generally regard as corrupt survey-based approaches

But each standard creates a problem:

  • Public office: what if the rules themselves are corrupt?
  • Public interest: who decides what counts as the “public interest”?
  • Public opinion: what if citizens disagree?

Classification following Heidenheimer (1970).

Nye’s public-office definition

Nye (1967): corruption is a deviation from the formal duties of a public role for private gain.

  • Strength: gives us a clear benchmark — what are the official duties and rules?
  • Limitation: it makes formal rules the standard for deciding what counts as corruption

But what if the rules themselves permit favoritism?

Legal ≠ necessarily honest

Lobbying, patronage, preferential access, or conflicts of interest may influence public decisions without violating formal duties.

Change the measure, change the picture

Bar chart of Corruption Perceptions Index 2024 scores for six Latin American countries, ranging from Costa Rica at 58 down to Venezuela at 10, with Mexico at 26 highlighted.

Bar chart of the share of public-service users who paid a bribe in the past year, 2019, for the same six countries, ranging from Costa Rica at 7 percent up to Venezuela at 50 percent, with Mexico at 34 percent highlighted.

So what? 26/100 on perception, 34% on experience — these do not contradict each other; they measure different things. “Corruption” is an umbrella: next we name the parts.

Part III: A Vocabulary of Corrupt Acts

Mapping the terrain: what does the actor do?

Three families of corrupt act, by what the actor does: exchange (bribery, extortion, influence trafficking, clientelism, patronage, vote buying), appropriation (embezzlement, fraud), and system manipulation (state capture).
  • These are mechanisms, not mutually exclusive boxes — one scandal can involve several at once

Author’s illustration. The individual terms are standard in the literature; the grouping by mechanism is the author’s.

What is being exchanged?

Money or favors → decisions

  • Bribery: giving a benefit to influence an official decision
  • Extortion: demanding a payment through threats or coercion
  • Influence trafficking: selling access or influence over a decision-maker

Benefits → political support

  • Clientelism: repeated benefits in exchange for political support
  • Patronage: public jobs in exchange for political loyalty
  • Vote buying: money or goods in exchange for a vote

The key question: what is the exchange?

  • Lobbying: trying to influence a decision through permitted advocacy
  • Influence trafficking: selling improper access or influence
  • Case 1 — mordida: Was the payment offered, demanded, or forced?
  • Case 2 — despensas: Were the goods given in exchange for votes?

Appropriation: embezzlement and fraud

  • Embezzlement: diverting resources you were entrusted to manage
  • Fraud: deception — phantom firms, fake invoices, ghost workers
  • Often no victim on the spot, so no complaints — detection needs audits
  • But fraud can be collusive too: fake bidders are an exchange
  • La Estafa Maestra combined diversion, deception, and collusion

Thomas Nast cartoon of the Tweed Ring standing in a circle, each man pointing at the next.

“Who stole the people’s money?” — everyone points elsewhere

Thomas Nast, Harper’s Weekly, 1871. Wikimedia Commons, public domain.

System manipulation: state capture

  • Firms or elites shape the rules themselves, not their application
  • Laws, decrees, court doctrine — written to order (Hellman, Jones, & Kaufmann, 2003)
  • Ordinary bribery bends existing rules; capture rewrites them
  • Once capture succeeds, the conduct it protects becomes legal
  • A clean example: a beverage association drafts the sugar-tax rules; congress adopts them verbatim

No bribe is needed. The rule itself is the payoff.

The four cases, classified

Designated summary slide — our vocabulary applied:

Case Best label Illegal? Entrusted power abused?
1. Police stop petty bribery / extortion yes yes — police discretion
2. Despensas vote buying / clientelism the practice: yes; the 2012 case: alleged yes — via party machine
3. Revolving door influence trafficking no anticipated, not proven
4. Estafa Maestra embezzlement + fraud yes yes — budget authority
  • Alleged, documented, unethical, illegal, convicted — different claims; keep them apart
  • One “no” entry in the Illegal? column — hold that thought

Exercise 2 — Name that act

≈ 5 minutes

Prompt. Classify each scenario with one vocabulary term; be ready to defend your choice:

  1. A mayor puts a campaign worker on the payroll as “special advisor”; he never appears at work.
  2. A customs officer charges $20 to process legal imports faster.
  3. A health official’s hidden firm wins the ambulance contract via fake bidders.
  4. A construction group funds the campaign of the legislator writing the procurement law; the final text exempts contracts like theirs.

Scenarios overlap — but some labels fit the defining mechanism better than others.

Why is this a problem?

The system can distort public decisions without anyone explicitly breaking the rules.

Thompson (2018): institutional corruption

Corruption can exist even without a corrupt individual.

The problem lies in the incentives built into the institution, not necessarily in a secret deal or guilty intent.

Definitions decide what we see

There is no neutral way to count corruption.

Today’s single idea, three consequences:

  • Definition → measurement: CPI counts perceptions; barometers count experiences
  • Definition → law: OECD-style offences reach bribes, not revolving doors
  • Definition → reform: punish individuals, or redesign institutions?
  • Legal corruption is invisible to every illegality-based measure
  • What your country counts shapes what it fixes

Conclusion — six things to take away

  1. Corruption is contested by design — four cases
  2. Definitions are choices — public office, entrusted power, or specified offences
  3. Three questions travel anywhere — who had power, how was it misused, who benefited
  4. The real fight is over “misused” — formal duty, public interest, or public opinion
  5. Three families of act — exchange, appropriation, system manipulation
  6. Some corruption is legal — and invisible to every illegality-based measure

Look back at your four votes from the opening: which one changed, and which definition changed it?

References

Animal Político & Mexicanos Contra la Corrupción y la Impunidad. (2017). La Estafa Maestra: Graduados en desaparecer dinero público. https://www.animalpolitico.com/estafa-maestra/

Heidenheimer, A. J. (Ed.). (1970). Political corruption: Readings in comparative analysis. Holt, Rinehart and Winston.

Hellman, J. S., Jones, G., & Kaufmann, D. (2003). Seize the state, seize the day: State capture and influence in transition economies. Journal of Comparative Economics, 31(4), 751–773. https://doi.org/10.1016/j.jce.2003.09.006

Nye, J. S. (1967). Corruption and political development: A cost-benefit analysis. American Political Science Review, 61(2), 417–427. https://doi.org/10.2307/1953254

OECD. (2008). Corruption: A glossary of international standards in criminal law. OECD Publishing.

References (continued)

Thompson, D. F. (2018). Theories of institutional corruption. Annual Review of Political Science, 21, 495–513. https://doi.org/10.1146/annurev-polisci-120117-110316

Transparency International. (2019). Global Corruption Barometer — Latin America & the Caribbean 2019: Citizens’ views and experiences of corruption. Transparency International.

Transparency International. (2025). Corruption Perceptions Index 2024. Transparency International. https://www.transparency.org/en/cpi/2024

United Nations. (2003). United Nations Convention against Corruption. United Nations Office on Drugs and Crime.

World Bank. (1997). Helping countries combat corruption: The role of the World Bank. World Bank.