2.3 · Why a decent person sells a vote

IntermediateDuration ~12 min readTools One municipality or barangay you actually know

The big idea: selling a vote is not a failure of character, it is a correct valuation of two promises, and the whole loop stays standing because the promise that never gets kept is the one made by the state.

The analogy

You have been offered the back end.

Small cash day rate now, or points on the picture. On paper the points are worth several times the cash, and the producer has the projections to prove it. Everyone in the room can do the arithmetic. And almost everyone takes the cash.

Not because they are short-sighted. Because they know what a promise from that producer is worth. He has made the same offer on four films and settled on none of them, there is no bond, no escrow, no completion guarantee, and no body that will hear you if the points never arrive. The cash is certain and it is today. The points are a number multiplied by the probability that a man who has never paid will pay, and that probability is not a mystery to anyone who has worked with him.

The crew member who takes the cash is not the problem in that room. He is the only person pricing the offer correctly.

Now notice what fixing it would take. You would not fix it with a speech to the crew about believing in deferred compensation. You would fix it with a bond, an escrow account, a union that will actually file, or a producer who has paid out before in a way anyone can check. You would change what stands behind the promise.

Hold that shape. It is the whole lesson.

How it works

Say the behaviour first, plainly, because you have already noticed it and easing into it would read as evasion.

People sell their votes. Pulse Asia found that around the 2016 national election 22 percent of Filipinos had been offered money or goods for a vote, and 76 percent of those offered took it. This is not a rumour, and it is not a few bad barangays.

You are allowed to hold whatever you already feel about that. Just set it next to the arithmetic and see which of the two is actually doing the explaining.

The arithmetic. A household with no savings, no insurance and no reliable public service is offered cash by a candidate who is going to win the seat whether or not this household participates. The cash is certain and it arrives this week. The alternative is a promise of better roads, better clinics and better schools from an institution that has not once delivered any of those within this family’s living memory. There is no third option on the table. Under those terms, taking the money is the highest-return decision available, and refusing it is a donation to an outcome that was already fixed.

That is the point where contempt runs out of work to do.

The framework for it is clientelism, and Allen Hicken gives it four defining elements: it is dyadic, it is hierarchical, it iterates, and it is contingent. Contingency is the load-bearing one. The benefit arrives if and only if the support does, and both parties know it. Strip contingency out and the same cash becomes an entitlement, which behaves completely differently. That single property is the difference between a public programme and a private arrangement wearing a public budget.

Now the loop, which is the analytical core of this level. Five steps, and each one is individually rational for the person taking it.

Poverty and vulnerability make small transfers decisive, because a peso is worth more to a household with nothing than to a household with something. That produces contingent loyalty, the vote-selling you just priced. Contingent loyalty produces entrenchment, because incumbents who control transfers convert them into vote share, and families that hold a seat long enough start holding several at once. Entrenchment produces extractive use of public resources, because family control of office plus family control of local business turns a public budget into private income, and because budgets flow to visible, divertible projects rather than to services whose payoff arrives after the next election. That produces weak public services, which leaves the household exactly where step one found it, still needing a patron.

Each arrow has evidence under it, and one of them is uncomfortable in a way worth stating out loud.

Julien Labonne studied the randomised rollout of the Pantawid Pamilyang Pilipino Program, the conditional cash transfer scheme, and found incumbent vote share substantially higher in municipalities where the programme covered every village than where it covered half. His own conclusion is the awkward one: a programme built to reduce vulnerability may also prevent local incumbents from ever being replaced. Reducing poverty and reducing patronage are not automatically the same project.

The second uncomfortable thing is one the course refuses to overstate. The strongest work on dynasties and poverty, by Ronald Mendoza and colleagues, supports the claim that poverty entrenches dynasties more firmly than the reverse, with less evidence that dynasties worsen poverty overall, though a worsening effect does show up in provinces outside Luzon. The dramatic version of that sentence is not what the identification supports, so the course says the weaker true thing. Later work adds the business channel, including a fat dynasty count and construction-sector links, and it is in the receipts.

The tools arrive now, because now they have something to explain.

Douglass North called institutions “the rules of the game in a society.” Organisations are the teams. If your team keeps losing, read the rulebook before you blame the players. Two Filipino crews shoot on the same Tuesday. One has a completion bond, a signed call sheet and an insurer. One has none of them. Same people, same skills, radically different behaviour by Friday. That is North in one shot, and it is the same shape as the back end.

Daron Acemoglu and James Robinson then split rulebooks into two kinds. Inclusive ones spread participation, apply law without favour, and deliver services that give everybody a usable field. Extractive ones concentrate power and wealth narrowly and produce no incentive to invest in anything whose return arrives later. Their explicit claim, the one this whole level is standing on, is that nations fail because of institutions rather than because of culture.

Which is why the loop is not a story about Filipino weakness. It is a rulebook, and the money trail through pork barrel spending and the flood control audit now running is what it looks like from above while the vote-selling is what it looks like from below. Same machine, two ends.

The pattern

Five steps, and each one is the sensible move for the person taking it. Vulnerability makes small transfers decisive. Decisive transfers buy contingent loyalty. Loyalty entrenches families. Entrenched families route budgets to what can be diverted. Diverted budgets leave services weak, which recreates the vulnerability. The loop is not held together by bad character. It is held together by the absence of any alternative at step one. Every arrow is a place to cut. Ask of any anti-corruption proposal you meet: which arrow does this attack, and what evidence says it moves.

The receipts (evidence, if you want it)

• This lesson is explainer-led on the framework and data-led on the arrows. No verified free video covers the loop as a whole. Every arrow below has a free source.
• The framework, free PDF. Allen Hicken (2011), “Clientelism,” Annual Review of Political Science, 14, 289 to 310. https://www.thepolicypractice.com/sites/default/files/2023-02/Clientelism.pdf . Four defining elements: dyadic relationships, contingency, hierarchy and iteration. Hicken also sets out why politicians target the poor, which is the standard diminishing-marginal-utility argument rather than a claim about the poor.
• Vulnerability and clientelism, free PDF. Gustavo J. Bobonis et al., “Vulnerability and Clientelism,” NBER Working Paper 23589. https://www.nber.org/system/files/working_papers/w23589/w23589.pdf .
• The scale of vote-buying. Pulse Asia found that 22 percent of Filipinos said they had been offered money or a material thing for their vote in the May 2016 elections, and that 76 percent of those offered accepted. Reported via Transparency International, “How much is your vote worth?”: https://www.transparency.org/en/news/how-much-is-your-vote-worth . Free. Separately, the Transparency International Global Corruption Barometer for Asia, fieldwork 2019 to 2020, records 28 percent of Filipinos reporting they had been offered a bribe in return for their vote. Note on the dates: some secondary reporting attaches a 2021 date to the Barometer figure. The course uses the fieldwork window rather than a publication year, and the Barometer should be pulled directly before it is spoken on camera.
• Arrow B, the cash transfer finding, free PDF. Julien Labonne (2013), “The local electoral impacts of conditional cash transfers: Evidence from a field experiment,” Journal of Development Economics. https://julienlabonne.wordpress.com/wp-content/uploads/2013/05/cct_elections.pdf . Incumbent vote share was 26 percentage points higher in municipalities where 4Ps covered all villages than where it covered half. Labonne’s own wording is that such programmes “might have nefarious long-term consequences by preventing replacements of local incumbents.”
• Arrow C, and the honesty point. Ronald U. Mendoza, Edsel L. Beja Jr., Victor S. Venida and David B. Yap (2016), “Political dynasties and poverty: measurement and evidence of linkages in the Philippines,” Oxford Development Studies, 44(2), 189 to 201. Free preprint: http://moda.addu.edu.ph/wp-content/uploads/2016/03/IPS-11_1-Political-Dynasties-and-Poverty_Evidence-from-the-Philippines.pdf . Using an instrumental variable strategy, the supported finding is that poverty entrenches dynasties, with less evidence that dynasties exacerbate poverty overall, and a worsening effect appearing in provinces outside Luzon. The strong version, “dynasties cause poverty,” is not what the paper supports and is not used here.
• The fat dynasty numbers. An Ateneo School of Government study covering 1988 to 2019 found the share of “fat” dynasties, families running several relatives in different offices at once, rose from 19 percent to 29 percent. Reported via Inquirer Business, “Philippine political dynasties: Boon or bane?”: https://business.inquirer.net/446032/philippine-political-dynasties-boon-or-bane . Mendoza, Yap et al. (2022), “Political dynasties, business, and poverty in the Philippines,” World Development Perspectives, open access: https://www.sciencedirect.com/science/article/pii/S2667319322000222 , adds that more than 60 dynasties have links to construction companies.
• The tools. Douglass C. North (1990), Institutions, Institutional Change and Economic Performance, Cambridge University Press: institutions are “the rules of the game in a society or, more formally, the humanly devised constraints that shape human interaction.” Paid; chapter 1 previews free at Cambridge Core. Daron Acemoglu and James A. Robinson (2012), Why Nations Fail, Crown, paid, with the free citable equivalent being the Nobel committee’s scientific background for the 2024 prize: https://www.nobelprize.org/uploads/2024/10/advanced-economicsciencesprize2024.pdf .
• The money trail, and two citations that are not yet closed. The Priority Development Assistance Fund scam was exposed by Nancy C. Carvajal in the Philippine Daily Inquirer in July 2013, and the Supreme Court struck down the congressional pork barrel as unconstitutional later that year in Belgica v. Ochoa. The case number and decision date have not been confirmed against the Supreme Court E-Library for this course, so they are not stated here. Separately, the Department of Public Works and Highways confirmed 421 ghost flood control projects out of roughly 8,000 audited, with Senate hearings and an Independent Commission for Infrastructure following. That story is live and unresolved while this course is being written, and its timeline needs confirmation before narration. Both are named rather than quietly used.
• What is deliberately not claimed. No average price of a bought vote, because the research base does not carry a defensible national figure. No claim that vote-selling determines national election outcomes, which is a much stronger claim than the evidence supports. And no claim that any individual voter you have in mind was acting rationally, because the loop is a statement about incentives facing households in aggregate, not a verdict on a person.
• The patronage loop diagram. One page, the five steps and the citation for each arrow. The Practice page carries the Level 2 capstone brief.

The patronage loop, one-page diagrampatronage-loop.pdf139 KBCourse original, formatted from this lesson's own sources

Your one move

Pick one place you actually know. The municipality your family is from, the barangay you live in, the town you shot in last year. Somewhere you can name real people and real projects.

Draw the loop for it. Five boxes down the page, in this order: vulnerability, contingent loyalty, entrenchment, extraction, weak services. Then fill each box with what you actually know about that specific place, not what you know about the Philippines in general.

You will not be able to fill all five. Mark the empty ones, because an empty box is the honest output here and it tells you what you would have to go and find out.

Then one line at the bottom, and this is the part that counts: which single arrow would you cut first, and what would you need to make the cut. Not which one is worst. Which one you could actually reach. An arrow you can reach with a camera, a database or a room full of filmmakers is worth more than an arrow that requires a constitutional amendment.

Thirty minutes. Done when five boxes exist, the empty ones are marked, and the last line names one arrow and one instrument.

Want to go further?

Do it a second time for the film industry instead of a municipality. Vulnerability is a crew with no savings between jobs. Contingent loyalty is who gets called back. Entrenchment is the small number of people who decide what gets made. Extraction and weak services you can work out yourself. It is the same diagram, and it is the one Level 3 spends seven lessons filling in.

The needle: the office you met in Lesson 1.2 was designed to be paid by the people it governed, which meant the state was never the thing that arrived. Four centuries later a household is still pricing two promises, and still getting the same answer. The voter is not the reason the loop turns. He is the person standing where it comes back around.

Terms introduced

Terms introduced in this lesson

• Clientelism
• Contingency
• Vote-buying
• Institutions (as rules of the game)
• Extractive and inclusive institutions
• Fat dynasty
• The patronage loop
• Pork barrel

Each term links to the glossary, where it is defined by the way you just used it.

Check

Check yourself

1. Hicken lists four defining elements of clientelism: dyadic relationships, contingency, hierarchy and iteration. Why is contingency the load-bearing one?

   Because it is what separates a clientelist exchange from a public policy. The benefit arrives if and only if the support does, and both sides know it. Remove contingency and the same transfer becomes an entitlement, which is a different machine entirelyBecause it is the element that makes the relationship illegal under Philippine lawBecause it is the newest of the four and therefore the least studiedBecause it is the only one that can be measured in a survey

2. Labonne studied the randomised rollout of the 4Ps cash transfer programme and found incumbent vote share far higher where the programme covered every village. What is the uncomfortable implication he himself drew?

   That a programme designed to reduce vulnerability can also entrench the person administering it, which he put as a risk of preventing replacements of local incumbents. Reducing poverty and reducing patronage are not automatically the same projectThat voters in covered villages were bribed directly with programme funds, which is why the vote share movedThat cash transfers do not reduce poverty and should be replaced with jobs programmesThat the programme was captured by fraud and most of the money never reached households

3. The strongest available identification on dynasties and poverty supports one direction better than the other. What does the course say, and why does it matter that it says the weaker thing?

   That dynasties cause poverty, which is the finding, and the course states it plainlyThat there is no relationship at all, and the dynasty material belongs in a different lessonThat the two are correlated and causation is unknowable, so the loop cannot be drawnThat poverty entrenches dynasties, with less evidence that dynasties worsen poverty overall, though a worsening effect does appear in provinces outside Luzon. It matters because a documentary that claims the strong version can be dismantled by anyone who reads the paper

4. A friend says people who sell their votes are simply too stupid or too greedy to want anything better. What does the arithmetic of the loop do to that claim?

   It refutes the claim by showing that most Filipinos refuse the money when offeredIt confirms the claim, but adds that the greed is understandable given the circumstancesIt shows the claim is offensive, and the honest reply is that judging people says more about the judge than the judgedIt leaves the behaviour exactly where it was, a household selling a vote, and moves what the behaviour is evidence of. Certain cash today against a public service that has never once arrived is a correct read of two promises, not a failure of character

Check answers

You can move on when you can… draw the five-step patronage loop, name the evidence behind each arrow, and say which single step you would attack first and why.

Mark this lesson complete

Go deeper

• Next up: 2.4 · The lazy Filipino, and who needed that story, which takes the oldest accusation against Filipinos, finds out who was making it and what they needed it for, and lets a Filipino answer it in his own words from 1890.
• The Culture research map carries the loop in full, drawn, with every arrow cited and every gap named. Part 2 is the analytical core of this level and it is published openly.
• Hicken (2011) is free, twenty pages, and the cleanest short statement of what clientelism is and is not. Read it if you want the framework without the Philippine case attached.
• Back to 1.6 · Two parties, one class, sixteen elections for the political half of this story. Patronage and the patron-client tie were taught there as history. This lesson is what they look like at the household.