287 | Jean-Paul Faguet on Institutions and the Legacy of History
What this covers
Jean-Paul Faguet and Sean Carroll examine how specific historical events shape contemporary societies through measurable institutional pathways. The conversation centers on two complementary case studies: Colombia, where the colonial encomienda—a brutal forced-labor system—inadvertently created incentives for durable local state institutions that persist and correlate with better modern development outcomes; and Bolivia, where an imported left-right party system collapsed because it never mapped onto the society's actual social cleavage. Faguet argues that the broad "extractive versus inclusive" institutional taxonomy misses crucial causal patterns visible only when institutions are studied at local and sectoral levels, showing how path dependence operates through concrete mechanisms rather than abstract labels.
The discussion moves across several distinct territories. Faguet explains how cleavage theory illuminates why Bolivia's party system survived decades of coups and hyperinflation before suddenly collapsing around 2003—the real cleavage was ethnic and geographic, not economic—and connects this pattern to current Western politics, where the dissolution of working-class identity is eroding left-right politics itself. He addresses the methodological challenges of isolating historical causation, describing how neighbor-pair comparisons and reconstructed conquistador routes serve as natural experiments. The conversation also examines why identical constitutional rules produce radically different outcomes across cultures, the structural vulnerabilities of first-past-the-post versus proportional representation systems, and how political sorting in the United States has eliminated the cross-party ideological overlap that once enabled institutional cooperation. Throughout, Faguet positions institutions within a power-institution feedback loop, treating economic and political power as inseparable forces rather than independent domains.
Faguet argues that specific historical institutions, not high-level 'extractive vs. inclusive' labels, leave measurable legacies: in Colombia the colonial encomienda indirectly produced stronger local state institutions and better modern development, while in Bolivia an imported left-right party system collapsed because it never matched the society's true ethnic cleavage.
- The encomienda's lasting benefit came not from forced labor itself but from the incentives it created to build durable local state institutions.
- Party systems survive or collapse depending on whether they map onto a society's real social cleavage rather than an imported one.
- Disaggregating institutions below the national level reveals causal patterns the macro extractive/inclusive framework misses.
Colonial institutions and historical contingencies produce persistent long-term effects on development and institutional capacity.
- To rule out that better development reflects pre-existing 'locational fundamentals' (good soil, water, trade routes) rather than the encomienda, the researchers use neighbor-pair fixed effects (comparing adjacent municipalities where one had encomienda and one did not, so unmeasured local characteristics cancel out) plus an instrumental variable—the least-cost path of the conquistadores reconstructed from a 1560 imperial agent's records and NASA geographic data—which correlates with encomienda presence but not directly with modern outcomes.
“We use a method called neighbor pair fixed effects... we're going to use a methodology that compares neighbor pairs where one did and the other did not have Encomienda.”
- Contrary to the flattering story Latin Americans tell themselves, the conquistadores were not great lords but the second, third, and fourth sons of minor nobility who had little to inherit; educated and with some means to finance the voyage, they sought to set themselves up as lords in the New World rather than remain marginal in Spain.
“the people who went were the second, third, and fourth sons of the minor nobility who didn't have that much to inherit because the first son was gonna inherit what there was”
- In areas without encomienda there was no incentive to build a colonial state, and pre-existing indigenous institutions collapsed because indigenous populations declined by over 95%, leaving scattered people without governing institutions; thus the advantage of encomienda areas was partly that the alternative areas had their institutions run into the ground.
“you have population declines in excess of 95%. So if 95% of the indigenous population has died, the institutions that they ran fall completely apart.”
- The cutting edge of the field treats colonialism as a 'complex treatment' that produced many effects, many bad and some horrific, but also some good incremental effects (such as infrastructure like railroads that lowered transport costs and enabled economic integration); recognizing the marginal benefit of an outcome is distinct from endorsing the immoral process that produced it.
“colonialism was a complex treatment that did a number of different things, many of them bad, some of them absolutely horrific, immoral, et cetera. But others had good effects, like the colonies without doubt built some infrastructure”
- From as early as 1539, encomienda-based local governments performed sophisticated state functions—providing potable water (e.g., digging canals to bring mountain water to a town fountain in Tunja), managing waste, and regulating commerce and food safety—constituting a more capable and involved local state than would be expected for the 1500s, especially in Latin America, even as the national imperial state remained chronically weak.
“it's the Encomienda based local government that's doing that. They manage waste, they regulate commerce.”
- Recent empirical work (Nunn, Wantchekon, and others) shows the harmful effects of slavery persist into the present: in West African regions with severe slave trafficking, people today trust each other less—linked to historical slave-raiding that pitted ethnic groups against one another—and education and health outcomes remain worse in places that suffered heavy slavery.
“in places where there was severe slave trafficking on the western coast of Africa, today people trust each other less”
- Roughly 170 Spaniards defeated indigenous forces of over 10,000 warriors in Colombia (and ~176 against ~30,000 in Peru) through the element of surprise, gunpowder, steel, and crucially the indigenous cosmography that interpreted the Spanish as godlike messengers who had to be obeyed—a historical outcome that remains a genuine mystery since the larger force could plausibly have destroyed them.
“they won because of the element of surprise, psychology. They had gunpowder and they had steel.”
Party systems collapse when organized around false cleavages rather than societies' actual economic or social divisions.
- Bolivia's political party system was remarkably stable from 1952 to 2002—the same parties and individuals returned after coups, 50,000% inflation, and roughly 15 coups d'etat—yet it suddenly collapsed around 2003, an event popularly but implausibly attributed to a few demonstrations in Cochabamba and La Paz that could not realistically topple a system that survived far larger shocks.
“The same political parties came back after coups, after macroeconomic shocks, after guerrilla uprisings.”
- Per Lipset and Rokkan's cleavage theory, a viable party system must map onto the main political cleavage that actually defines a society; Bolivia's pre-2003 left-versus-right system was built on a false cleavage because Bolivia never had an industrial revolution and thus had no self-identified working class and few capitalists, so the system collapsed and was replaced by parties organized around the real cleavage—ethnic, identitarian, and geographic identity.
“Bolivia had kind of the wrong political party system built on a false cleavage that did not define the Bolivian society”
- Left-right politics is fundamentally economic and positive-sum: a growing pie lets capitalists and workers both gain, so compromise solutions are always possible; but if politics shifts to identitarian lines (ethnicity, religion, language), it becomes zero-sum—one group is on top or the other is—making agreeable solutions very difficult and threatening political stability. This dynamic, first visible in institutionally weak Bolivia, is now appearing across the West.
“the thing about left-right politics is that it's always possible to find a positive, some solution that makes everyone happy, because you can expand the pie”
Identical formal rules produce radically different institutional outcomes across cultural and economic contexts.
- Institutions written in identical black-and-white rules (e.g., constitutions copied from the US by Latin American republics, or shared French/British administrative structures left to African colonies) produce radically different outcomes when transferred across cultural contexts, demonstrating that formal rules alone do not determine institutional behavior.
“the Latin American republics come up a little bit later in the early 1800s and they basically copy large amounts of it and these institutions in Latin America do not work the same way that they work in the US, even though articles are copied almost verbatim”
- Beyond economic determinism, culture and human psychological attitudes—acquired from society and fellow human beings—influence how institutions develop and whether they succeed, which is why identical formal rules produce different outcomes in different cultural contexts.
“it's not purely economic determinism, that human beings have attitudes that we get from psychology, from our other fellow human beings in a society, and those also help influence the development of institutions and their success”
Presidentialism without ideological party overlap creates brittle systems lacking escape valves for resolving conflict.
- Per Adam Przeworski's comparative research, the American-style system of presidentialism and separation of powers has only really worked in the United States; the many countries that copied it (all of Latin America) failed, because parliamentary systems have an escape valve—a vote of no confidence triggering an election—that presidential systems lack, allowing a hobbled party to stagger on while resentment builds, and the historic ideological overlap between US parties was part of what made the exceptional American system work.
“the American-style system of presidentialism and separation of powers only really works in the United States of America. Many, many, many countries have copied this... And it doesn't work anywhere except for the US.”
- Through most of the 20th century, US parties overlapped ideologically—many Southern Democrats were right-wing and many Northeastern Republicans were left-wing, belonging to a party for tribal or historical rather than ideological reasons—but over time, as visualized through roll-call votes, they separated completely so there is now no overlap, producing en-masse party-line voting and removing the cross-party common ground that once enabled cooperation.
“there's enormous overlap amongst the parties all through the 20th century... many Southern Democrats were actually fairly right-wing people, and many liberal Republicans from the Northeast were actually fairly left-wing people”
- First-past-the-post voting turns relatively small electoral majorities into large legislative majorities, producing strong but brittle governments that can enact sweeping change (a 35% vote can yield a vast parliamentary majority—an 'electoral dictatorship'), whereas proportional representation gives parties seats proportional to votes, better representing sincere preferences and enabling coalitions but producing only minor reshuffles; PR is better when things go well, but first-past-the-post's capacity for a clean sweep is valuable when things go badly.
“first-past-the-post takes relatively small electoral majorities and turns them into big legislative majorities”
- Beyond ideological separation, one of the two US parties has been taken over by people who appear to want to degrade democratic institutions and break down checks and balances—something that has never happened before in American history, where parties fought viciously but still respected the institutions and accepted losing—making a polarized system in which one side systematically tries to change the rules to favor itself genuinely dangerous.
“one of the two political parties has been taken over by frankly crazy people as far as I can tell, many of whom seem to want to degrade the institutions of democracy itself”
Social scientists lack robust general knowledge of when and why good governance arises across different contexts.
- Despite decades of research, social scientists do not really know when, why, or how good governance arises; we have only highly contingent hypotheses with scattered evidence for particular cases, and what works in one place and time (modern Germany) failed in the same place at another time (Germany 1890s-1940s) and offers little guidance for the rest of the world.
“the simple one sentence answer is nobody really knows. Sometimes it goes well, and very often it goes badly”
- The dominant 'extractive versus inclusive' framing of national institutions is too high a level of aggregation; institutions vary substantially below the national level (Mississippi differs from Massachusetts) and across dimensions (legal system, policing, electoral norms, party systems), so studying specific institutions across space yields different and more accurate answers.
“We think that's just like way too high level of aggregation in multiple ways in terms of space institutions that are not only national”
- Although Colombia became independent in 1821 and has been independent over 200 years, the Colombian state still has not fully occupied its territory: in places in 2024 there is no police station, mayor's office, school, or hospital, even though people live there and self-organize, reaching only about 350 of roughly 1,120 municipalities historically.
“there are places today in 2024 where the Colombian state has yet to arrive in the sense that there's no police station, no local mayor's office, no school and no hospital”
- Venezuela went from being arguably the richest, most sophisticated, and most politically stable country in Latin America—with ~90% voter registration and turnout—into a declining autocracy after Hugo Chávez's '21st century socialism' gutted its industrial sector, leaving an oil-dependent economy where the government pays off the military, elites tied to the government prosper, everyone else works informally, and about a third of the population has fled.
“Venezuela was the most sophisticated, the richest, and the most politically stable country in Latin America”
- The traditional European left is umbilically tied to unions and a self-identified working class, but that identity is dissolving: people work hard but no longer see themselves as working class (the gig-economy Uber driver thinks of himself as an entrepreneur rather than a factory worker among thousands), so left-right politics is in steep decline across the West for reasons strikingly similar to Bolivia's.
“people work hard in Britain and Germany and France today, but they don't see themselves as a working class. That's not how they conceive of themselves.”
- Electoral fraud is constrained by the size of the true margin: it is feasible to steal an election you lose narrowly (55/45), but when an autocrat loses by a large margin (e.g., the Venezuelan opposition's claimed ~70/30), the population directly knows the result in their own communities, making the claim to have won far less sustainable—as Pinochet discovered in the 1988 Chilean plebiscite when independent observers and parallel vote counts made his 60/40 loss impossible to hide.
“It is one thing to cheat, if you're losing 55/45. It's something else to cheat, if you're losing 70/30 and having lost 70, like the people know”
- Bolivia's decentralization created local governments that previously did not exist, generating a new class of politicians competing for local positions; with about a ten-year lag these politicians realized they need not join elite parties led by wealthy white La Paz residents and could form their own parties based on the society's real ethnic-geographic cleavage, enabling the rise of Evo Morales.
“largely because of decentralization, and when you created all these local governments, you got a new class of politician competing for local government jobs, positions that you didn't have before”
Institutions are fundamentally expressions of economic power applied to politics, feeding back into economic dominance.
- Following a classical political economy tradition (which Faguet characterizes as a revived, non-Leninist Marxism), institutions are at some level the expression of economic power applied to politics, which then feeds back into economic power in a continuous cycle; treating economic and political spheres as distinct, as much North American political economy does, risks missing the key dynamic of how the world works.
“they're just showing that institutions are at some level the expression of economic power applied to politics, which then feeds back into economic power and the two is a continuous cycle”
- The encomienda, a brutal colonial forced-labor institution, is statistically tied to better development outcomes today across about 20 measures (GDP per capita, wages, health, education, infant mortality, literacy, and institutional capacity) in Colombia—not because forced labor itself helped, but because the presence of encomienda gave Spanish lords strong incentives to build durable local institutions to protect their property rights, and those local institutions persisted and grew in capacity within a chronically weak imperial state.
“The presence of Encomienda gave these lords a very strong incentive to set up local institutions to protect their property rights, one of which, one of the key ones of which was the Encomienda.”