The Political Economy of Latin America

Deficit-financed redistribution, riding a narrow commodity rent (the money oil or copper earns over and above what it costs to get out of the ground) until the foreign exchange (the dollars a country needs to pay for its imports) runs out — that machine, not ideology, is Latin America's recurring tragedy. The region's current drama is a natural experiment in the four ways a country can try to break it.

Most coverage of the region reaches for a pendulum. A Pink Tide of leftist governments — the wave of left-wing presidents elected across South America from about 2000 — rises in the 2000s, a right-wing reaction follows, then another turn. The labels are real enough. They explain almost nothing. Argentina ran the identical deficit-financed boom under Peronist and non-Peronist administrations alike; Chile has alternated between left and right four times since 2005 without ever losing its fiscal rule (a standing law that caps what the government may spend, whatever the copper price does); Nayib Bukele governs El Salvador in a style that is unmistakably populist while running an IMF-supervised fiscal consolidation (deficit reduction carried out to a schedule the Fund monitors). What actually determines a country's trajectory is duller and more durable than a banner colour: the fiscal-monetary-exchange-rate regime — how a country taxes and spends, who controls the printing of its money, and how it sets the price of its currency — and whether anything can stop a president from breaking it.

This primer builds that working model. It teaches the mechanism first, then uses it to read every country in the region — including the four that now command attention, each of which is a different attempted exit from the same machine.

Latin America and the Caribbean is roughly a US$7 trillion economy growing at about 2.3 per cent a year and, on the Economic Commission for Latin America and the Caribbean's own reading, stuck there — four consecutive years at approximately that rate, which the Commission calls "a low-growth-capacity trap."^{[CEPAL BP-2025]} It holds roughly a third of the world's copper reserves and, on identified-resource measures, over four-tenths of the world's lithium. It has just produced, within eighteen months, the sharpest fiscal turn in Argentina's modern history, the most extreme incarceration regime on the planet, and the removal of a sitting head of state by United States military force.

§0The central insight, stated once

Latin America's political economy is organised around capturing and spending a commodity rent that the state does not control the price of; macroeconomic populism — growth and redistribution now, with the risks of deficits, inflation and the foreign-exchange constraint discounted, as §2 sets out exactly — is the political technology for spending that rent faster than it arrives; and the region's institutions have historically been too weak to stop it — so the question of the moment is not who is left or right, but which of the four available exits from that machine actually holds.

Everything that follows is an attempt to make the reader earn that sentence rather than accept it. Part I builds the engine: where the rent comes from, how the populist cycle converts it into a crisis, what institutional features let the cycle repeat, and which external forces set its tempo. Part II places every country in the region on the map the engine produces, with depth on the four live experiments. Part III asks whether any of it lasts.

NoteA correction to the brief, made on the evidence

This report was commissioned to treat "the October 2026 midterm elections" as Argentina's forthcoming referendum on Javier Milei's adjustment. That election has already happened. Argentina voted on 26 October 2025, electing half the Chamber of Deputies and a third of the Senate; La Libertad Avanza took roughly 40.8 per cent of the national vote against about 25–31 per cent for the Peronist opposition, and winning in sixteen provinces including Buenos Aires; contemporaneous reporting put its haul at 64 of the 127 contested lower-house seats and 14 of the 24 contested Senate seats.^{[AS/COA; Americas Quarterly; NPR]} The Buenos Aires Herald, writing two days before the vote, described the 26 October ballot as renewing exactly those 127 lower-house and 24 Senate seats.^{[Buenos Aires Herald, 24 October 2025]} Argentina renews its Congress biennially, so there is no national election in October 2026; the next is the general election of October 2027. That changes the question §6 has to answer: the referendum has been held and won, so Argentina's stabilisation is no longer being tested by an election but by something harder — the reserve position, and time.

Part IHow it works — the engine beneath the politics

§1The rent base: what the region actually sells, and who gets paid

Start with the balance of paymentsThe running tally of every payment that crosses a country's border: what it earns from exports, tourism, remittances and foreign investment, against what it pays out for imports, interest and profits repatriated by foreign owners. Domestic policy is settled in the local currency. The balance of payments is settled in dollars, and only one of those can be printed., because that is where every Latin American crisis eventually gets settled. A country can run any domestic policy it likes for as long as it can pay for its imports. And it pays for them out of its foreign exchange reserves: the stock of dollars and gold the central bank holds. Reserves are the buffer between a policy and its consequences. While they last, a government can hold its currency at a price the market does not believe and sell dollars to importers below cost; when they run out, the adjustment that was being postponed arrives all at once. The region's ability to pay for imports rests, to a degree that has barely changed in a century, on a handful of commodities whose prices are set somewhere else.

Two very different economies make the point. Guyana, the region's newest petro-state, earned 81.1 per cent of its 2024 goods exports from a single Harmonised System chapterThe internationally standardised customs classification in which traded goods are coded. At its broadest ‘chapter’ level it sorts all merchandise into fewer than a hundred boxes — mineral fuels are one chapter, ores another. It is the unit in which export dependence becomes visible. — mineral fuels.^{[UN Comtrade]} Brazil, the region's most diversified large economy and a genuine industrial power, still drew 53.3 per cent of its 2024 exports from five primary chapters: fuels, oilseeds, ores, meat and sugar.

Guyana is what a pure rent economy looks like: one product, one operator, one price. Brazil is what the region's best case looks like after eighty years of deliberate industrialisation: a real manufacturing base, a world-class services sector, and an export book still more than half primary. If Brazil cannot escape commodity dependence, no one in the region escapes it by accident.

Rent is doing specific work here, and the report leans on it repeatedly. It needs a definition.

Rent is not the same as revenue, and it is not the same as pricing power. Chile's copper miners have almost no ability to set the copper price; they earn rent anyway, because Chilean ore bodies sit far down the global cost curveRank every mine in the world by what it costs to produce a tonne, cheapest first, and that ranking is the cost curve. Producers at the low end still earn a profit at prices that put the high end out of business — which is how a country can earn a rent from a commodity whose price it has no power to set.. Venezuela's Orinoco heavy crude is expensive to lift and upgrade; the rent there was always thinner than the reserve headline suggested, which is one reason the country's collapse ran faster than most observers expected. Where the rent is fat and the extraction is capital-intensive, the state can tax it without visibly taxing voters. That is the fiscal fact that shapes Latin American politics.

1.1Where the rent physically sits

Two maps matter, and neither is drawn on paper. The first is hydrocarbons.

As recently as 2013 Venezuela produced 2.5 million barrels a day of crude and condensateThe light liquid hydrocarbon that comes up with natural gas and turns to liquid at the surface. Oil statistics usually count it alongside crude because it is refined the same way; ‘crude and condensate’ is simply the honest label for the total., and Guyana produced none.^{[EIA International Energy Statistics]} By 2025 Venezuela was at 973,000 barrels a day and Guyana at 747,000 — and Guyana is still ramping while Venezuela is still repairing. Brazil, meanwhile, produced 3.77 million barrels a day, more than the next two producers combined. The region's oil centre of gravity has moved decisively south and east, out of the Caribbean basin and into the South Atlantic. That single fact reorders a great deal of regional politics, and Part III returns to it.

The second map is metals, and the concentration there is extraordinary.

Chile and Peru between them hold 27.1 per cent of the world's identified copper reserves and produced 34.8 per cent of the world's mined copper in 2025 — 5.3 million and 2.7 million tonnes respectively out of a world total of 23 million.^{[USGS MCS 2026, copper]} Add Mexico and Latin America holds roughly a third of the world's copper reserve base and 37.8 per cent of its mine output. There is no substitute for copper in electrification at anything like current cost, and the identified reserve base is not being replaced as fast as it is consumed. Nothing else the region owns gives it a claim on the twenty-first century like that one.

Lithium is the other. On the United States Geological Survey's identified-resource measure, Argentina holds 28 million tonnes, Bolivia 23 million and Chile 13 million, out of roughly 150 million tonnes worldwide — 42.7 per cent between them.^{[USGS MCS 2026, lithium]} But resources are not reserves, and the difference is the entire story of the lithium triangle. On the reserves measure — the economically extractable subset — Chile has 9.2 million tonnes and Argentina 4.4 million, while Bolivia has none listed at all. In production, Chile made 56,000 tonnes in 2025 and Argentina 23,000; Bolivia does not appear in the USGS producer table. Bolivia holds the largest identified lithium resource on earth and sells essentially none of it.

That gap — enormous resource, no production — is this report's argument in one country. Geology is not destiny. Institutions convert resource into rent, and Bolivia's have not.

1.2The chain from ore to budget — and where the state actually sits

A commodity endowment becomes a political fact by stages, and the popular version compresses them into one.

Politics is fought over the wrong layers. Layers 1 and 5 are what it talks about — sovereignty over the resource, and what the money buys. Layer 4 is what determines whether the country survives the next price cycle, and it is almost never the subject of an election. A government that nationalises an oil field (layer 1) while abolishing its stabilisation fundA pot the state pays commodity revenue into when the price is high and draws on when it is low, under a written rule about how much may be taken out. Its purpose is not investment returns. It is to stop a windfall turning into a permanent spending commitment. (layer 4) has made itself poorer and more fragile while appearing to have made itself stronger. Venezuela did exactly that. So, in a milder form, did Bolivia.

§2Macroeconomic populism: the Dornbusch–Edwards cycle, step by step

In 1989 Rudiger Dornbusch and Sebastián Edwards wrote down what they had watched happen twice within their own careers — in Salvador Allende's Chile and Alan García's Peru — and gave it a name. Their definition is exact, and the exactness matters: the looseness of the word "populism" in ordinary use has cost a great deal of analytical clarity since.

The authors were explicit that they were not making a conservative argument. Their stated purpose was "not a righteous assertion of conservative economics, but rather a warning that populist policies do ultimately fail" — and fail in a specific, repeatable pattern that ends with the intended beneficiaries poorer than when it began. That last clause is the whole point, and it is why the framework has survived four decades of ideological weather.

The paper's four phases are the spine of this primer.

Phase I works. This is the part that most commentary gets wrong, usually by skipping it. The initial expansion genuinely delivers. Output rises, real wages rise, employment rises, and inflation does not appear — because idle capacity absorbs the demand, because inventories are run down, and because imports financed by reserve decumulation fill the gap. Dornbusch and Edwards are blunt: in phase one "the policy makers are fully vindicated in their diagnosis and prescription." The programme is not a con. It is a loan against the future taken out in a currency the borrower cannot print.

Phase II is where the balance of payments starts to bite. Bottlenecks appear as domestic demand outruns domestic supply and foreign exchange grows scarce. Now the government faces its first genuine choice: devalue and let real wages fall, or hold the exchange rate and ration the foreign exchange. Almost every populist programme chooses to hold, because devaluation is inflationary and visibly cuts living standards. So price controls arrive, then exchange controls, then import licensing. Inflation rises anyway. Wages are indexedWages, contracts or tax brackets tied by formula to last period's inflation, so they rise automatically as prices do. Indexation protects incomes for a while, and it makes inflation far harder to stop, because yesterday's price rise is already wired into tomorrow's wage bill. to keep up. And the budget deficit, as the original paper puts it, "worsens tremendously as a result of pervasive subsidies on wage goods and foreign exchange" — because holding an overvalued exchange rate is a subsidy, paid out of reserves, to everyone who buys dollars at the official rate.

Phase III is the collapse. Shortages become pervasive, inflation accelerates sharply, the foreign exchange gap becomes undeniable, and capital flees. The fiscal position deteriorates violently from both ends at once: tax collection falls in real terms because of the lag between accrual and payment under high inflation — the Olivera-Tanzi effect — while subsidy costs explode. The government belatedly tries to stabilise by cutting subsidies and devaluing. Real wages fall hard. Politics becomes unstable.

Phase IV is orthodox stabilisation under new management, usually with an IMF programme, and it ends with real wages "significantly lower than when the whole episode began." Dornbusch and Edwards give the reason in one line that repays memorising: "capital is mobile across borders, but labor is not." Capital exits during phase III at the official exchange rate. Labour cannot. The adjustment therefore falls on wages, and it falls persistently, because the episode has destroyed the investment that would have raised productivity later.

Populism does not fail because it is generous. It fails because generosity financed by reserves is a loan the poorest end up repaying.

The Dornbusch–Edwards result, restated

2.1Why the exchange rate is always the tell

Watch the exchange rate regime, not the rhetoric. Every episode of macroeconomic populism converges on an overvalued, administratively defended exchange rate, because that is the only instrument that simultaneously suppresses measured inflation, raises measured real wages, and cheapens imports. Overvalued means something precise: the government has fixed the official price of a dollar below what a buyer would actually pay for one. Anybody who can obtain dollars at that price is being handed a subsidy, and anybody who has to earn them — the exporter — is being taxed. Defending the rate means selling reserves to whoever turns up, which is why the policy's shelf life is measured in reserves rather than in years. It is the sweetest of all short-run policies and the most expensive of all medium-run ones.

The overvalued rate does four things at once, and only the first is visible to voters. It raises real purchasing power in imported goods. It destroys the profitability of tradable productionAnything a country makes that could be sold abroad, or that competes at home against imports — crops, ore, manufactures. Non-tradables are the rest: haircuts, housing, government. An overvalued currency squeezes the tradable side, which is precisely the side that earns the dollars., so the export base shrinks precisely when export earnings are most needed. It creates an arbitrage between the official and parallel ratesThe black-market exchange rate: what a dollar actually fetches when the official price is fixed too low. The gap between the two is the clearest single measure of how far a currency's official price has drifted from what people will pay. that becomes, in short order, the country's most profitable industry — and a mechanism for transferring public reserves to the politically connected. And it converts every subsequent correction into a discrete devaluation shock rather than a gradual adjustment.

Argentina's December 2023 devaluation produced a 25.5 per cent monthly consumer price rise in that single month.^{[INDEC via datos.gob.ar]} Venezuela's multi-tiered exchange regime between 2003 and 2018 produced parallel-market premia measured in orders of magnitude. Bolivia's peg, defended from 2011 until it could not be, has produced a parallel rate and a chronic dollar shortage since early 2023. These are not three stories. They are one story at three points on the same curve.

2.2Two distinctions the reader must hold

The first is between populism as an economic programme and populism as a political style. They are separable and they cross-cut. Milei is a stylistic populist of the purest kind — anti-institutional, plebiscitaryGoverning by direct appeal to the people over the heads of parliaments, courts and parties, with the leader's mandate treated as a standing referendum. It describes where a government locates its authority, and says nothing at all about its economics., rhetorically incendiary — running the most orthodox macroeconomic programme in the hemisphere. Bukele is likewise a stylistic populist running an IMF-supervised consolidation. Meanwhile the Kirchner administrations in Argentina and the Chávez and Maduro administrations in Venezuela were economic populists of the textbook kind, and Chávez in particular combined that with plebiscitary style. The combinations are four, not two, and confusing them produces bad predictions.

The second is between a structural stabilisation and a terms-of-tradeThe price of what a country sells, set against the price of what it buys. When copper or soy rises faster than the cost of a country's imports its terms of trade improve and it is better off without having done anything differently — and the reverse happens just as passively. sugar high. When a commodity price doubles, a rent-dependent economy grows, its fiscal balance improves and its currency strengthens — regardless of policy. Distinguishing this from a genuine regime change requires asking what the fiscal position would look like at the long-run price, which is precisely what a structural-balance rule is designed to force a government to compute. Chile does that computation by law. Almost no one else does.

§3The institutional soil: why it recurs

A cycle that failed once would be a mistake. A cycle that has run repeatedly across a dozen countries for seventy years is a structural feature, and it demands a structural explanation. What the explanation has to account for is the absence of a constraint on the executive — some person or body, outside the president's own office, able to say no and make it stick. A court that rules against him. A legislature that can refuse a budget, a central bank that can decline to print, an electorate that can remove him. Where none of those bites, nothing at all stands between a commodity windfall and a spending decision, and that is the second of the two variables this report reads every government by. Four features of Latin American governance do most of the work.

PresidentialismA system in which the head of government is elected directly by voters for a fixed term, separately from the legislature — as against a parliamentary system, where the executive holds office only for as long as the legislature supports it. A president cannot easily be removed between elections. That is a source of stability and of impunity in equal measure. with weak legislative counterweight. Almost every state in the region runs a directly elected executive with substantial decree power and a legislature elected on a different clock. When the president's coalition is weak, the result is gridlock and governance by decree; when it is strong, there is very little to stop the executive from doing whatever it wishes with the fiscal accounts. Both failure modes are visible right now — Peru has had nine presidents in a decade, four of the last six elected presidents removed by Congress on grounds of "permanent moral incapacity";^{[CNN; Al Jazeera; AP via NPR]} El Salvador has a legislature that has renewed a state of exception monthly for four years without meaningful debate.

Central banks that are formally independent and practically not. The mechanism that converts a fiscal deficit into inflation is monetary financing, and the institution that is supposed to refuse is the central bank. Monetary financing is the shortest route from a spending decision to a price rise. A government that cannot borrow the money and will not tax for it asks its own central bank to credit the Treasury's account instead: new money, created against nothing, chasing the same quantity of goods. Nobody votes for the inflation that follows, which is precisely why it is the favoured way to pay for things nobody will vote to fund. In the region's crisis cases the central bank has not refused. Cuba's 2025 deficit was financed principally by the issuance of sovereign bonds absorbed domestically, which the Economic Commission for Latin America and the Caribbean notes plainly "supposes an increase in the monetary base."^{[CEPAL, Cuba country note 2025]} That is monetary financing described in polite language. Argentina's central bank financed the Treasury directly for most of two decades. Where central bank independence is real — Chile, Brazil, Peru, Mexico, Uruguay, Colombia — hyperinflation does not occur, and that correlation is close to perfect in the region.

Fiscal rules that exist on paper. Paraguay's Fiscal Responsibility Law, Chile's structural-balance rule, Guyana's Natural Resource Fund withdrawal formula and Brazil's successive spending caps are all genuine attempts to bind the future. Their track records differ sharply, and the difference is instructive: rules survive when breaking them carries a cost the executive cannot absorb, and are suspended when they do not. Paraguay's government has been reducing its deficit from 4.1 per cent of GDP in 2023 to a targeted 1.9 per cent in 2025 and 1.5 per cent in 2026, in line with the upper bound the law sets.^{[CEPAL, Paraguay country note 2025]} Brazil, with a nominally binding framework, ran a deficit above 7 per cent of GDP in 2025 and carries gross general government debt of 93.3 per cent of GDP.^{[IMF WEO April 2026]}

Party systems that do not transmit commitment across administrations. This is the least discussed and possibly the most important. A fiscal anchor is whatever obliges a government to keep its spending tied to what it can actually finance: a deficit ceiling, a balanced-budget law, a rule that computes the budget at a long-run commodity price rather than today's. It is not a target. A target is what a government says it will do; an anchor is what happens to it if it does not, and that distinction does most of the work in the pages that follow. It is also a promise about the future, and a promise is only worth what the institution making it can enforce after the person making it leaves. Where parties are durable, programmatic and capable of disciplining their own presidents — Chile until recently, Uruguay throughout — the anchor survives alternation. Where the vehicle is a personal movement built for one election, nothing survives its founder. Milei's La Libertad Avanza did not exist in 2020. Bukele's Nuevas Ideas did not exist in 2017. Neither has yet demonstrated that it is a party rather than a candidate.

That cuts against the enthusiasm currently attached to both men. Inference A reform that depends on a single leader is not an institution; it is a hostage. The test is not whether the reformer succeeds. It is whether his successor is constrained to continue.

§4The external tempo: dollars, China, and the commodity clock

None of the above is generated inside Latin America. The region is a price-takerA seller too small or too replaceable to move the price it receives, and which therefore takes whatever the market posts. The opposite is a price-maker. Almost every Latin American exporter is the former, which is why so much of the region's income is decided elsewhere. in three markets it does not participate in setting, and the interaction of those three markets sets the tempo of the domestic cycle.

The dollar cycle and United States monetary policy. Latin American external debt is predominantly dollar-denominated, and Latin American commodity exports are dollar-priced. When the Federal Reserve tightens and the dollar strengthens, the region gets hit twice: debt service rises in local-currency terms exactly as commodity export receipts fall. This is why so many Latin American crises cluster around United States tightening cycles — 1982, 1994, 2018 — and it is the single most reliable external predictor of regional stress. The mechanism is not mysterious. It is a currency mismatchOwing money in one currency while earning it in another. It works quietly until the currency you owe in becomes more expensive, at which point the debt grows in real terms without anybody having borrowed a further cent. on a national scale.

Chinese demand. China's industrialisation was the proximate cause of the 2003–2013 commodity super-cycleA commodity price upswing that runs for a decade or more, driven by a structural shift in demand rather than by the ordinary boom and bust of inventories. Long enough that a whole political generation mistakes it for the new normal. that funded the Pink Tide. The correlation that followed is often mistaken for a policy result. Left-wing governments were elected across South America between 2000 and 2006 and presided over a decade of rising real wages and falling poverty. So did the region's right-wing governments over the same period. The common factor was the price of soy, copper and oil, not the ideology of the administration. When the super-cycle ended in 2013–2014, the entire cohort ran into difficulty simultaneously, regardless of banner. Correlation was doing nearly all of the explanatory work.

Chinese finance is now a different animal from Chinese demand. The era of enormous oil-backed policy-bank loansLending by state-owned banks whose mandate is their government's foreign policy rather than a commercial return. The money is real; the pricing is strategic, and the terms are rarely public. — the China Development Bank and Export-Import Bank of China commitments that reached tens of billions of dollars a year in the early 2010s, and which CSIS puts at more than US$60bn to Venezuela alone over recent decades^[CSIS] — is over. By 2022 new CDB and Eximbank commitments to the entire region had fallen to US$813m,^{[Inter-American Dialogue / BU GDPC]} and the lending that has resumed is smaller, more targeted, and directed at transport and industrial projects rather than at sovereign budget support. The strategic implication is that China's leverage in the region has shifted from creditor to customer, which is a weaker but more durable form of influence, and one much harder for Washington to counter.

Put the three together and the region's rhythm becomes legible. A commodity upswing plus loose global money finances a domestic expansion; the expansion outlives the upswing; the tightening cycle arrives and forces the adjustment. The countries that survive it are the ones that saved during the upswing, and the countries that saved are the ones with a rule strong enough to make saving politically survivable. Everything else is detail.

Part IIWhy it matters — the great divergence, and every country on the map

§5Four paths out of the same machine

The same engine, running in twenty countries with different fiscal institutions and different endowments, has produced four distinct destinations — and a country's destination is predicted far better by its fiscal anchor and the constraint on its executive than by anything it calls itself.

Before placing anyone, here is the region's macroeconomic state as the International Monetary Fund saw it in April 2026. This is one consistently constructed dataset across all twenty-two economies for which it exists, which makes it the right basis for comparison; national accounts differ in scope from country to country and are used later for depth, never for cross-country ranking.

Three things in that table need flagging first. Cuba is absent, and the absence is a fact rather than an omission: Cuba is not a member of the Fund and does not appear in the World Economic Outlook database at all. Its numbers in this report come from the Economic Commission for Latin America and the Caribbean and from Cuban official statements, and they are treated with correspondingly lower confidence.

Second, Venezuela's figures are internally strange — 1.5 per cent growth in 2025 on IMF staff estimates against 6.5 per cent on the Economic Commission's, and average inflation of 252 per cent in 2025 rising to a projected 387 per cent in 2026, against the Commission's last published reading of 23.6 per cent as of October 2024.^{[CEPAL BP-2025]} These are not small discrepancies; they are order-of-magnitude disagreements between two multilateral institutions about the same country in the same year. The reason is that the Commission works largely from Venezuelan official series and the Fund from staff reconstruction. Known unknown Any confident statement about Venezuela's current growth or inflation rate is overstated, and this report treats Venezuelan macro data as directionally usable and numerically unreliable.

Third, Bolivia. A general government deficit of 11.6 per cent of GDP in 2025, output contracting 1.2 per cent in 2025 and a projected 3.3 per cent in 2026, inflation near 20 per cent. That is a country in phase III, and §10 treats it as the region's live demonstration that the cycle has not been retired.

5.1The map

The horizontal axis is a real, sourced number: the general government fiscal balance in 2025. The vertical axis is an ordinal judgement, and the report states its coding rule rather than pretending to precision it does not have. A score of 3 means competitive elections, a judiciary that has ruled against the executive, and a legislature capable of blocking it; 2 means two of the three; 1 means one; 0 means none. Chile, Brazil and Colombia score 3 on the evidence of alternation and adverse court rulings; Argentina scores 2 (competitive elections and an independent judiciary, against a legislature the executive has repeatedly bypassed by decree); El Salvador scores 1 (elections are held, the courts and legislature are controlled); Venezuela scores 0 on the evidence of the disputed 2024 presidential result and the current absence of any electoral timetable.

Read the quadrant as four destinations rather than four rankings.

5.2The pendulum that is not a pendulum

Set the ideological label of each government against the thing that actually determines outcomes.

The text in each cell is the government's ideological alignment. The colour is something else entirely: whether the executive faced a binding constraint. The rows move; the colours mostly do not. Brazil, Chile and Colombia have swung repeatedly between left and right while remaining dark throughout — alternation without regime change. Nicaragua and Venezuela went red and stayed red irrespective of what happened elsewhere. El Salvador is the only row that has moved from dark to red in the period, and it did so under a government that is, on the fiscal numbers, more orthodox than the one it replaced.

Ideological alternation and institutional erosion are independent variables. A country can swing hard left and hard right without losing anything that matters, and a country can hold an election every five years while dismantling every check that makes the election meaningful. Coverage that tracks the first and ignores the second will consistently mis-forecast the region.

The 2025–2026 electoral sequence reads differently in that light. Bolivia elected Rodrigo Paz in October 2025 with 54.5 per cent in a run-off, ending twenty years of Movimiento al Socialismo government.^[PBS/AP] Chile elected José Antonio Kast with 58.17 per cent in December 2025.^{[Servel via Emol/Meganoticias]} Honduras's electoral council declared Nasry Asfura president-elect in December 2025.^{[CNE via HRN]} Costa Rica elected Laura Fernández with 48.3 per cent in February 2026.^{[TSE via La Nación]} Peru elected Keiko Fujimori by about 49,600 votes in June 2026 and inaugurated her on 28 July.^{[ONPE via Reuters; AP via NPR]} Colombia elected Abelardo de la Espriella on 21 June 2026 with 49.66 per cent against Iván Cepeda's 48.70.^{[Registraduría via Reuters]}

Six consecutive right-of-centre wins looks like a regional turn, and every one of those campaigns was fought on crime and cost of living rather than on ideology. But it is anti-incumbency more than it is conservatism. The Pink Tide of 2000–2010 was likewise a wave of anti-incumbent votes, cast against governments that had presided over the 1990s adjustments. What is being observed is ordinary democratic accountability in economies whose living standards are set by a commodity price they do not control. Voters punish whoever is in office when the terms of trade turn. That will happen again to the current cohort, and the useful question is what each of them will have built by the time it does.

§6Argentina: the adjustment turnaround, and what the monthly data actually says

Argentina has achieved the hardest thing in the Dornbusch–Edwards sequence — a genuine phase IV that has held for two and a half years — and the monthly data show that it did so with a serious relapse in the middle that the headline story has quietly dropped.

Javier Milei took office on 10 December 2023 inheriting the classic terminal position: a general government deficit of 5.3 per cent of GDP, an exchange rate held far above its market-clearing level by a dense apparatus of controls known as the cepo, foreign exchange reserves that were negative on a net basis, and inflation running at an annualised triple-digit rate.^{[IMF WEO April 2026]} He devalued immediately and cut spending immediately. The first month's consumer price index rose 25.5 per cent.^{[INDEC via datos.gob.ar]}

6.1The fiscal swing, measured honestly

Argentina's general government went from a deficit of 5.3 per cent of GDP in 2023 to a surplus of 0.5 per cent in 2024 — a swing of 5.8 points of GDP in one year, achieved without a debt restructuring and without a banking crisis. There is no comparable peacetime fiscal adjustment in the region's recent history.

Two comparability warnings, both material, and both routinely ignored in commentary. The first: the figures above are general government on the Fund's definition, consolidating the provinces. The Argentine government's own headline figures are federal cash, and they are different numbers describing a different thing. A second vocabulary problem sits inside the same sentences. A primary balance sets what the government takes in against everything it spends except the interest on its debt; the overall balance puts the interest back in. The gap between the two is the price of the past, and it is why the same year can be a surplus on one measure and a deficit on the other without anybody having lied. On the federal cash basis the primary surplus was 1.8 per cent of GDP in 2024 and 1.4 per cent in 2025, with an overall cash surplus of 0.2 per cent in 2025; the Ministry of Economy reported a 2025 financial surplus of AR$1,453,819 million, or 0.2 per cent of GDP, on a primary surplus of AR$11,769,219 million, or 1.4 per cent.^{[Ministerio de Economía / argentina.gob.ar]} These are not contradictory; they are different perimetersWhich entities a fiscal number actually covers. A federal perimeter is the central government alone; a general-government perimeter adds the provinces, the municipalities and the social security funds. Two honest accountants working to different perimeters will publish different deficits for the same country in the same year.. Quoting one against the other is the most common analytical error made about Argentina.

The second warning is where the general-government number earns its keep. The Fund's 2026 Article IV consultationThe IMF's annual health check on a member country, named for the article of the Fund's charter that requires it. Staff visit, examine the accounts and publish an assessment. It happens whether or not the country is borrowing, which is what makes it the closest thing to an independent audit of a national economy. records that "at the provincial level… primary balances deteriorated by nearly 1 percent of GDP in 2025, mainly due to an expansion in real spending amid easing borrowing constraints."^{[IMF Country Report, Argentina 2026 Art. IV / Second Review]} That is the gap between the federal surplus and the general-government deficit, and it identifies the adjustment's structural weak point precisely: as soon as the provinces regained market access, they started spending again. A fiscal anchor that binds only the federal government in a federal country is an anchor with a known leak.

6.2Disinflation — and the relapse the headline hides

The shape of the line matters more than its endpoints. The standard account of Argentina runs "inflation fell from 25 per cent a month to under 2 per cent." True at the endpoints, and misleading in between. Monthly inflation fell to 1.5 per cent in May 2025 and then rose for seven of the next ten months, reaching 3.38 per cent in March 2026, before falling again to 1.89 per cent in June 2026.^{[INDEC, IPC junio 2026]} The twelve-month rate at June 2026 was 33.5 per cent, on a year-to-date accumulation of 16.8 per cent.

The Fund's staff report names the cause without euphemism: "political uncertainties ahead of the October midterm elections led to a surge in resident dollarization and reserve losses, triggering much tighter monetary conditions," and the end-2025 net international reservesReserves counted after subtracting what the central bank has itself borrowed or pledged — the dollars it genuinely owns rather than the dollars merely sitting in its account. It is the stricter measure, and the one Fund programmes are written against, because a gross figure can be flattered by borrowed money. target was consequently missed.^{[IMF Art. IV 2026]} That is a textbook phase-II episode inside a phase-IV programme: the exchange rate came under pressure, the central bank lost reserves, and prices reflected it with a lag. It resolved. It resolved because the government won the election and then, from January 2026, changed the monetary and foreign-exchange framework — the exchange-rate band's crawlA pre-announced rate at which a currency's official price is allowed to slide, a little at a time, rather than being held fixed and then broken in a single shock. Re-indexing the crawl means changing the formula that sets that speed. was re-indexed to lagged INDEC inflation^[BCRA] and a daily reserve-purchase programme was introduced, under which the central bank bought roughly US$7.5bn in the first months of 2026.

Two things follow, and they point in opposite directions. The stabilisation demonstrably survived a stress test, which is genuine information about its robustness. And the stress test was an ordinary domestic election, which tells you how thin the buffer still is. Both are true. A reader who takes only the first is being sold the story; a reader who takes only the second is missing the achievement.

6.3Poverty: the number, and the 40 per cent of it that is arithmetic

Argentine poverty rose above 50 per cent in the first half of 2024 — the Fund's staff put the peak at 52.9 per cent — and fell below 30 per cent by the third quarter of 2025, with INDEC's second-half 2025 reading at 28.2 per cent.^{[IMF Art. IV 2026][INDEC via Chequeado]} Real per-capita incomes in the bottom quintile rose about 33 per cent between the third quarter of 2023 and the third quarter of 2025, against 26 per cent in the top quintile. That is a real, large, progressive improvement, and it happened under a government that cut public spending hard. It is the most surprising fact in the Argentine story.

It also needs decomposing, and the Fund does the decomposition. Of the fall in the poverty headcount, roughly 40 per cent — 7.3 percentage points — is attributable to what the staff call the relative inflation effect: the basic consumption basket, weighted towards food, rose 270 per cent over the period while headline prices rose 307 per cent, mechanically lifting households across a poverty line that is defined by that basket. Labour income accounts for about 42 per cent (7.5 points), pensions about 2 points, and better-targeted transfers most of the rest.^{[IMF Art. IV 2026, Box 1]}

This is neither a debunking nor an endorsement. It is calibration. About three-fifths of the improvement is income — durable, if growth holds. About two-fifths is a one-time repricing of the basket against which poverty is measured, and it cannot recur, because the relative price move that produced it has already happened. Anyone forecasting further sharp poverty declines from the same mechanism is forecasting something the mechanism cannot deliver.

6.4The programme, the reserves, and the real test

Argentina is operating under a 48-month Extended Fund FacilityThe IMF's longest-tenor standard loan, built for a country whose problem is structural rather than a passing cash squeeze. The money is released in tranches against periodic reviews, and a missed target stops the next tranche. The conditions are the product; the cash is what enforces them. approved on 11 April 2025 with access of about US$21bn — 479 per cent of quotaEach member's subscription to the IMF, set by the size of its economy, which fixes both its voting share and how much it may normally borrow. Access far above 100 per cent of quota is exceptional and has to be approved as such. — of which roughly US$15.8bn had been disbursed by the second review, completed on 21 May 2026.^{[IMF Art. IV 2026 / Press Release]} The Fund projects growth of about 3.5 per cent in 2026 with end-period inflation around 25 per cent, and the programme targets net international reserve accumulation of at least US$8bn during 2026. Federal government debt is projected to fall from 84.7 per cent of GDP in 2024 to 73.2 per cent in 2026.

The fiscal anchor is real and has now survived two budget cycles and one election. The disinflationA fall in the rate at which prices are rising. Prices still go up, only more slowly. It is not deflation, where prices actually fall, and confusing the two is the commonest way to misread a stabilisation. is real and has survived one relapse. The reserve position is the weak point, has been the weak point since the beginning, and remains the weak point: the end-2025 target was missed, and the Fund's own language is that "reserve buffers [are] thin" and the programme "continues to be subject to elevated risks." Argentina's history is not a history of governments that failed to cut spending. It is a history of stabilisations that ran out of dollars.

Against that, one thing has changed structurally that has nothing to do with Milei. Argentine crude and condensate output rose from a 2018 trough of 490,000 barrels a day to 794,000 in 2025, and total liquids production passed 1.04 million barrels a day by early 2026, driven by the Vaca MuertaThe shale formation in western Argentina that holds the country's principal unconventional oil and gas play. Shale is produced by fracturing dense rock rather than tapping a conventional reservoir, which makes output a function of continuous drilling activity rather than of a single discovery. shale.^[EIA] An energy export surplus of a scale Argentina has not had in twenty years is arriving in the same window as the stabilisation. Distinguishing the policy from the endowment will take another three years of data, and honest analysis should say so now rather than after the fact.

§7El Salvador: the security-state turnaround, and the price on the receipt

El Salvador achieved the largest recorded fall in a homicide rate anywhere in the world in peacetime, and it did so by imprisoning roughly one adult in fifty — a trade whose terms are unusually explicit and whose exportability is the real question.

The achievement is enormous, and dismissing it is the fastest way to misunderstand Central America. El Salvador's homicide rate peaked at approximately 105 per 100,000 people in 2015, a level that had no peer outside active war zones.^{[HRW World Report 2026]} The Congressional Research Service, working from Salvadoran police data, records 53.1 per 100,000 in 2018 and 1.9 per 100,000 in 2024.^{[CRS IN12510]} El Salvador's Ministry of Public Security and Justice reported 82 homicides in 2025 for a rate of 1.3 per 100,000, with 1,102 days without a violent death recorded since 2019, 988 of them under the state of exception.^{[Ministerio de Seguridad Pública y Justicia]}

Two caveats that a serious reader must carry. The Salvadoran government uses a narrower homicide definition than most comparators, excluding killings by police, deaths in prison custody, deaths of alleged criminals at the hands of citizens, and human remains recovered from clandestine graves; the Congressional Research Service notes analysts' estimates that including those categories would raise the 2023 figure by up to 47 per cent.^{[CRS IN12510]} And access to the underlying data has been progressively restricted. Adjust for both and the fall is still one of the largest ever recorded. A rate of 1.9 becomes perhaps 2.8; against 105, the distinction is academic.

7.1The mechanism, and the bill

The instrument is a régimen de excepción — a state of exception — approved by the Legislative Assembly on 27 March 2022 for thirty days and renewed monthly ever since. It suspends the right to be informed of the reason for arrest, the right to counsel, and the limit on pre-trial detention.^[WOLA]

That last point comes from Human Rights Watch's own account: previous Salvadoran governments obtained lower homicide numbers by granting gangs "looser prison regimes, reduced sentences, early releases, and protection from extradition, in return for lowering homicides and political support during elections."^{[HRW World Report 2026]} El Salvador did not move from rule of law to a bargain. It moved from a covert bargain with criminal organisations to an overt suspension of law administered by the state. Whether that is an improvement is a real question with a defensible answer on both sides — and it is a different question from the one usually asked.

The costs are documented and severe. Amnesty International's July 2026 report concludes that the documented patterns of mass arbitrary detention, torture, enforced disappearance and deaths in custody "could constitute crimes against humanity under Article 7 of the Rome Statute."^{[Amnesty International]} Serving and retired police described arrest quotas, verbal instructions without documentary record, and profiling by neighbourhood and appearance. The Legislative Assembly has extended pre-trial detention for alleged members of criminal organisations to five years and authorised transferring children detained on organised-crime charges into adult prison facilities.

The exchange had a price, and both sides of it are on the record. Roughly 6,000 lives a year were being lost to homicide at the 2015 rate in a country of six million. The current rate saves nearly all of them. The cost is a permanent suspension of due process, a prison population at double capacity, several hundred deaths in custody, and — the part that will matter longest — a judiciary and legislature that no longer function as constraints on the executive at all. A reader who thinks that trade is obviously wrong should confront the 105; a reader who thinks it is obviously right should confront the 470.

7.2The economy underneath: dollarisation, remittances, and a walked-back experiment

El Salvador's macroeconomics are stranger and more constrained than its politics suggest. The country abandoned its own currency in 2001 and uses the United States dollar. That removes monetary populism from the menu entirely: there is no central bank to finance the Treasury, so the classic Dornbusch–Edwards inflation channel is simply unavailable. What remains available is debt, and El Salvador used it. Gross public debt has sat between 83 and 95 per cent of GDP since 2020 and stood at 87.2 per cent in 2025.^{[IMF WEO April 2026]}

The external account rests on remittances to a degree that has few parallels anywhere.

El Salvador received US$9,987.9m in remittances in 2025, up 17.8 per cent on 2024 and equal to roughly a quarter of GDP; the Northern TriangleGuatemala, Honduras and El Salvador taken together — the three Central American countries whose emigration, remittances and gang violence are conventionally analysed as one system. total was US$47,730.2m against US$39,732.9m in 2024.^{[IOM via EFE/Infobae; IADB]} This is the country's largest source of foreign exchange, exceeding exports, foreign direct investment and tourism combined. It is also, functionally, the reason dollarisation has been survivable: a permanent, price-insensitive dollar inflow substitutes for the reserve accumulation a central bank would otherwise have to manage.

§15 develops what follows from that. Remittances to the Northern Triangle rose 20 per cent in 2025, to US$47,730.2m, in the same year that migration to the United States collapsed to the lowest levels on record. Those two facts are usually reported separately. They are almost certainly the same fact.

On the Bitcoin experiment, the record is now clear and the correction was substantial. In February 2025, ahead of the Fund programme, the Bitcoin Law was amended to make private-sector acceptance voluntary, to repeal the obligation on public and private actors to accept it, to end payment of taxes in Bitcoin, and to remove its characterisation as a currency. The Fund's programme documents confirm the commitments: acceptance made voluntary, taxes payable only in United States dollars, government purchases of Bitcoin prohibited for the duration of the programme, and public-sector engagement in Bitcoin-related activity confined.^{[IMF Country Report 25/58]} The Fund's own assessment of the episode is deflating in its plainness: "Bitcoin use remains marginal… The use of Bitcoin as means of payment has been minimal, reflecting Bitcoin's high price volatility and limited trust in the technology."

El Salvador entered a 40-month Extended Fund Facility on 26 February 2025 with access of SDRSpecial Drawing Rights: the IMF's own unit of account, a basket of major currencies in which the Fund keeps its books and denominates its lending. Amounts are converted to dollars at the prevailing basket rate, which is why Fund figures are usually quoted twice. 1,033.92m, about US$1.4bn, under which the primary balance is to improve by 3.5 per cent of GDP over three years.^{[IMF Country Report 25/58]} The Economic Commission's reading is that the non-financial public sectorThe government plus its state-owned enterprises, with the central bank and the public banks left out. It is the wider view of what the state actually spends, and it catches a loss-making state oil or power company that a narrow budget line would miss. deficit including pensions fell from 4.5 per cent of GDP in 2024 to about 3.0 per cent in 2025 on current-spending restraint, with growth of 3.5 per cent.^{[CEPAL, El Salvador country note 2025]}

The composite is a government stylistically populist, fiscally orthodox under supervision, monetarily disarmed by dollarisation, externally dependent on a remittance flow it does not control, and institutionally unconstrained. An unusual configuration, and it puts the standard warnings about El Salvador on the wrong risk. The danger is not that Bukele will run a macroeconomic populist programme — he cannot, because there is no printing press and there is a Fund programme. The danger is that a country with no institutional constraint on its executive has no mechanism to correct a mistake, and that the security model's returns are already asymptotic while its costs continue to accrue.

§8Venezuela: collapse, capture, and the question nobody can yet answer

Venezuela is the completed version of the cycle — the only Latin American economy to lose roughly three-quarters of its output in peacetime — and the live question after January 2026 is not whether chavismo fell but whether it is reconstituting itself under new management with American sponsorship.

8.1The magnitude, first

Numbers this large lose their meaning without anchoring, so take them slowly. On IMF data, Venezuelan real GDP contracted every year from 2014 to 2020: −3.9, −6.2, −17.0, −15.7, −19.7, −27.7 and −30.0 per cent. Compounded, that is a cumulative real contraction of 74.4 per cent — the economy of 2020 was a quarter the size of the economy of 2013.^{[IMF WEO April 2026]} In current dollars, GDP fell from US$372.6bn in 2012 to US$42.8bn in 2020, a decline of 88.5 per cent. The United States lost about 27 per cent of real output in the Great Depression. Venezuela lost close to three times that, without a war.

The oil sector tells the same story in physical units, and physical units cannot be fudged.

Venezuelan crude and condensate output ran at 3.12 million barrels a day in 1998, the year before Hugo Chávez took office, and 2.5 million in 2013, the year he died. By 2020 it was 527,000 — a fall of 79 per cent from the 2013 level.^{[EIA International Energy Statistics]} The country with the world's largest claimed proved oil reserves could not produce enough gasoline for its own drivers.

Guyana produced its first commercial barrel in December 2019 and reached 747,000 barrels a day in 2025. A country of some 800,000 people now produces about three-quarters of what its 28-million-person neighbour produces, with an offshore development sanctioned in a decade rather than a state oil company built over seventy years. That is the clearest natural experiment the region offers on the difference between an endowment and the institution that develops it.

8.2How it happened — the mechanism, not the morality tale

Venezuela is not a story of socialism failing, or at least that description explains nothing that the general mechanism does not explain better. It is the Dornbusch–Edwards cycle run at maximum amplitude because the rent was unusually large and the constraint unusually absent.

Phase I (2004–2008) was funded by an oil price that went from roughly US$25 to nearly US$150 a barrel. Social spending rose sharply, poverty fell sharply, and the government was vindicated. This part was real and is often airbrushed out of hostile accounts. Phase II (2008–2014) brought the exchange controls: a multi-tier official rate, an import licensing regime, and price controls on food and basic goods. The parallel-market premium became enormous, and access to dollars at the official rate became the most lucrative business in the country — the mechanism by which public reserves were privatised. Simultaneously, PDVSA, the state oil company, was converted from an operator into a fiscal and social-programme vehicle, staffed after the 2002–2003 strike by political appointment. Capital expenditure fell and so, with a lag of several years, did production. Phase III (2014–2020) arrived when the oil price halved. Output was already falling; now revenue collapsed on top of it. The result was hyperinflation, the emigration of what is now over 6.9 million people, and a state that stopped delivering electricity and water.^[R4V]

United States sanctions, tightened materially from 2017 and extended to PDVSA in 2019, are real and made the trough deeper. They are not the cause. Output was already down 41 per cent from its 2013 level by 2018, before the PDVSA sanctions took effect, and the collapse in capital spending that produced that decline began years earlier. Attributing the collapse to sanctions requires ignoring the sequence. Attributing nothing to sanctions requires ignoring the depth of the trough. Both errors are common; the evidence supports neither.

8.3What actually happened in 2026

On 3 January 2026, United States forces struck military infrastructure in Venezuela and captured Nicolás Maduro and his wife Cilia Flores in Caracas; they were removed from the country and transported to New York to face narcotics and weapons charges.^{[CSIS; PBS NewsHour; CNN]} President Trump stated that the United States would be "running the country until such time as a just and orderly transition can take place."^[CSIS] Venezuela's Supreme Tribunal directed Vice-President Delcy Rodríguez to assume the powers of acting president, and Defence Minister Vladimir Padrino López supported the appointment. On 8 January the government announced releases of political prisoners; Foro Penal confirmed 383 released as of February, including Juan Pablo Guanipa, Freddy Superlano and Perkins Rocha.^{[NPR; NBC News; PBS]}

On 29 January the National Assembly approved, and Rodríguez signed, the most significant reform of the Organic Hydrocarbons Law since the 1976 nationalisation: it ends PDVSA's monopoly on primary activities, allows private companies to conduct upstreamThe exploration and production end of the oil business: finding the oil and lifting it. Downstream is refining and selling it. Who is permitted to operate upstream decides who carries the geological risk and who captures the barrel. operations and commercialise hydrocarbons directly, lowers the state's minimum joint-venture stake, caps royaltiesA payment to the resource's owner — here the state — calculated on what is produced rather than on what is earned. Royalties are collected whether or not the operator makes a profit, which makes them dependable for a treasury and unforgiving for a marginal field. at 30 per cent of extracted volumes and replaces the layered tax structure with a single integrated hydrocarbons tax capped at 15 per cent.^{[Mayer Brown; Baker McKenzie; Hogan Lovells; S&P Global]} The same day, the United States Office of Foreign Assets Control issued General LicensesBlanket permissions issued by the United States Treasury's sanctions office that unlock a whole class of otherwise-prohibited transactions for everyone who qualifies, without each firm having to apply. They are the fastest switch an administration has for turning a sanctioned economy back on. They can be revoked just as fast. 46 and 47 authorising established United States entities to trade Venezuelan oil and export diluentsLight hydrocarbons blended into extra-heavy crude so that it will flow through a pipeline and onto a tanker. The heaviest grades will not move without them, which makes access to diluents a physical constraint on output and not merely a commercial one.; GL 48 followed on 10 February, and GL 52 on 18 March broadly authorised transactions with PDVSA including new investment and joint-venture formation.^{[Mayer Brown; Morgan Lewis; Cleary Gottlieb; Holland & Knight]}

On 5 March the State Department announced that "the United States and Venezuela's interim authorities have agreed to re-establish diplomatic and consular relations," describing "a phased process that creates the conditions for a peaceful transition to a democratically elected government."^{[U.S. Department of State]} Embassy Caracas formally resumed operations, with Ambassador Laura F. Dogu having arrived in January as Chargé d'Affaires.^{[U.S. Department of State]}

Production has responded. Venezuela produced 973,000 barrels a day of crude and condensate in 2025 on EIA data; monthly figures reported for 2026 have run in the 1.0–1.1 million barrel a day range, with Chevron's joint ventures alone at roughly 280,000 barrels a day gross by the second quarter.^{[Reuters/industry reporting; see limits]} Estimated Monthly 2026 production figures here are reported estimates, not verified against an opened OPEC or EIA primary; the OPEC Monthly Oil Market Report was not retrievable this session and the EIA's international series for Venezuela runs to annual 2025.

8.4Transition or adaptation? The evidence on both sides

The question is genuinely open.

Inference The strongest single predictor of whether a rent-state democratises is not who holds the presidency but whether the rent flows through a rule or through a person. Venezuela's hydrocarbons reform of 29 January genuinely changes the rent's plumbing — private operators, a capped royalty, an integrated tax — and that is a more consequential democratising force than any personnel change, because it makes the state's income legible, contractual and, crucially, dependent on counterparties who will not invest without enforceable law. Foreign capital cannot be expropriated twice cheaply; the second time, nobody comes.

But the same reform, arriving before any electoral timetable, also describes a state that has learned to fund itself from foreign operators without needing to answer to voters. That is the Gulf model, not the Chilean one, and it has a long track record of stability without democracy. Which of the two Venezuela becomes will be settled by whether an electoral calendar is published, and by whom.

Two consequences follow at one remove. The first is migratory: 6.9 million Venezuelans are living outside the country, and the direction of that flow is now a policy variable rather than a constant. A credible transition reverses some of it, which would relieve Colombia, Peru, Chile, Brazil and the United States simultaneously — and would remove the single largest driver of anti-migrant politics across South America. A failed one restarts it. The second is Cuban, and it is treated next, because Venezuela's collapse as a patron is the largest external shock the Cuban system has faced since 1991.

§9Cuba: why the regime persists while the economy does not

Cuba is the region's control experiment: a state that removed the market entirely, never faced the populist cycle in its classic form, and is nonetheless failing through the same channel — a rent it did not earn, withdrawn by a patron it could not control.

Cuba does not appear in the IMF's World Economic Outlook because it is not a Fund member. It reports selectively. What follows is therefore assembled from the Economic Commission for Latin America and the Caribbean, which works from Cuban official series, and from Cuban government statements, and it carries lower confidence than any other country section in this report.

The Commission's December 2025 assessment was a contraction of 1.5 per cent in 2025 following 1.1 per cent in 2024, inflation of 14.95 per cent year-on-year in November 2025 (down from 27.0 per cent a year earlier), a fiscal deficit falling towards 6.6 per cent of GDP from 7.3 per cent, and a current accountThe part of the balance of payments covering trade in goods and services plus income and transfers such as remittances — everything except purchases and sales of assets. A current account deficit means a country is consuming more than it earns abroad, and somebody has to finance the difference. still in deficit on weak goods exports, falling tourism and slower remittances.^{[CEPAL, Cuba country note 2025]} Cuban officials speaking in mid-2026 have described a 5 per cent decline in 2025 and roughly 15 per cent cumulative contraction since 2020.^{[Reuters/UPI reporting, July 2026]} Those two accounts are not reconcilable and the report does not attempt to reconcile them; the divergence is itself the finding.

9.1The mechanism — fiscal dominance without a market

The Commission's description of how Cuba finances itself is the populist mechanism in a non-market setting. Total financing demand for 2025 was projected at 129,374 million pesos, of which 88,539 million (68.4 per cent) was the fiscal deficit and 29,531 million the amortisation of sovereign bonds. The principal source of funding was the issuance of further sovereign bonds, which, in the Commission's words, "supposes an increase in the monetary base."^{[CEPAL, Cuba country note 2025]}

That is monetary financing of a fiscal deficit — the identical engine that produced Argentine and Venezuelan inflation, running inside a command economy. The reason it produces shortages rather than measured hyperinflation is that prices are administered, so the excess purchasing power shows up as queues and empty shelves rather than as a price index. The suppression is cosmetic. The Cuban peso's informal exchange rate and the collapse of the ration system are where the pressure actually appears.

9.2The Venezuelan shock

Cuba's external position rested for two decades on subsidised Venezuelan crude — supplied on concessional terms in exchange for medical services — and on tourism. Both have gone.

Cuba suffered a total national grid collapse on 10 July 2026 — its second in five days and the fourth of the year — and another in early August.^{[CNN; Rigzone/Bloomberg]} President Miguel Díaz-Canel has attributed the crisis to fuel starvation and to United States pressure, stating that Cuba received a single fuel tanker over seven months, and that about 1,400 megawatts of distributed generation is unavailable daily as a result.^{[UPI; U.S. News/Reuters]} Attributed These figures are the Cuban government's own and have not been independently verified.

The National Assembly approved a package described as 176 economic and social transformation measures, of which 120 are said to be ready for implementation: easing restrictions on land use, allowing enterprises to import fuel and medicines, and permitting private tourism operators. Díaz-Canel stated that key sectors would not be privatised.^{[Reuters via U.S. News, 30 July 2026]}

9.3Why the regime survives anyway — and the exportable lesson

Economic failure does not, by itself, topple regimes, and the reason generalises well beyond Cuba. What topples regimes is the loss of the coercive apparatus's loyalty, and that is a function of whether the apparatus is paid and whether it believes it has anywhere to go. Cuba's security and military establishment holds significant economic assets directly, which means regime survival and personal solvency are the same problem for the people with the guns. Emigration, meanwhile, functions as a pressure-release valve that removes exactly the young, mobile, discontented population that would otherwise organise.

Contrast Venezuela, where the same two mechanisms operated for a decade and the regime nonetheless fell — not to internal pressure but to external force. The comparison sharpens a conclusion this report will restate in Part III: Observed across the region's authoritarian cases, no incumbent regime has been removed by economic collapse alone in the past twenty-five years. Nicaragua's economy grew 3.5 per cent in 2025 and its regime tightened; Cuba's has contracted for years and its regime has held; Venezuela's collapsed by three-quarters and its regime survived until an aircraft carrier group arrived. Anyone forecasting political change in Havana or Managua from an economic series is using the wrong instrument.

§10South America: the rest of the continent, located on the map

One figure makes the whole regional structure visible at once. It rates each country's endowments by how load-bearing they are for its external earnings and its budget — not by how large they are in absolute terms.

Each red cell marks a country whose external solvency rides on a single price it does not set. That is where the populist cycle finds its fuel and where the next crisis will originate. The pattern also shows why the region does not have one political economy but three: a hydrocarbon south-and-Caribbean, a metals spine down the Andes, and a remittance-and-services isthmus whose macro is effectively an appendix to the United States labour market.

10.1Brazil — the anchor economy with the loosest anchor

Brazil is the region's largest economy at US$2.28trn in 2025 and the one whose fiscal position is least consistent with its institutional strength.^{[IMF WEO April 2026]} It grew 2.3 per cent in 2025 with 5.0 per cent inflation, a general government deficit of 8.1 per cent of GDP and gross public debt of 93.3 per cent. The Economic Commission expects growth to slow to 2.0 per cent in 2026 on contractionary monetary policy and reduced fiscal impulse.^{[CEPAL BP-2025]}

Brazil has everything the framework says should protect a country: a genuinely independent central bank that has been willing to hold policy rates punishingly high, a competitive party system, a Supreme Federal Tribunal that has repeatedly ruled against sitting presidents, and functioning federalism. And it runs a deficit that would be a crisis anywhere else in the region. The reason it is not a crisis is that Brazil's debt is overwhelmingly domestic, in its own currency, held by a deep domestic institutional investor base — the one structural feature no other country in the region possesses. That converts the fiscal problem from a solvency question into a growth question: high real interest rates crowd outWhen government borrowing absorbs the available savings and pushes up the interest rate everyone else must pay, private investment that would otherwise have happened does not. The state does not forbid it. It simply outbids it. investment, so Brazil grows at 2 per cent instead of 4.

Economically, Brazil is now an energy and agriculture superpower. It produced 3.77 million barrels a day of crude and condensate in 2025 — more than any other country in the hemisphere except the United States and Canada — and mineral fuels were the largest single chapter of its 2024 goods exports at 17.0 per cent, ahead of oilseeds at 13.0 per cent.^{[EIA][UN Comtrade]} The country most associated with soy is now, by export value, an oil economy first.

Politically, Lula da Silva is standing for a fourth term in the general election scheduled for October 2026, having returned to office in 2022 by 1.8 points. Jair Bolsonaro is serving a sentence of 27 years and three months following his conviction, and is effectively removed from electoral politics.^{[Rio Times; ISPI; GlobalSecurity]} Recency caution Brazil's October 2026 election had not occurred as of this report's evidence cutoff, and nothing here should be read as a forecast of it.

10.2Mexico — the shadow over the isthmus

Mexico is outside the mandate's formal country list but cannot be omitted, because Central America's macro is largely a function of it and of the United States. Mexico grew 0.6 per cent in 2025 on IMF data and 0.4 per cent on the Economic Commission's, with inflation at 3.8 per cent inside Banco de México's target band, a general government deficit of 4.9 per cent of GDP and debt at 61.8 per cent.^{[IMF WEO][CEPAL, Mexico country note 2025]} The Commission attributes the slowdown to weaker consumption and contracting investment, aggravated by United States tariffs on some Mexican products.

Mexico's crude output has fallen from 3.07 million barrels a day in 1998 to 1.73 million in 2025, a managed decline at Pemex that no administration has reversed.^[EIA] Remittances fell in 2025 for the first time in over a decade: Banco de México recorded US$56.469bn over the first eleven months, 5.1 per cent below the same period of 2024, with the real purchasing power of those dollars down 14.4 per cent year on year in November as the peso appreciated 10.2 per cent. The first half of 2026 recovered 3.1 per cent to US$30.759bn.^{[Banco de México via Mexico Business News]} President Claudia Sheinbaum's government faces the United States–Mexico–Canada AgreementThe North American trade agreement governing Mexican and Canadian access to the United States market, successor to NAFTA. It obliges its three members to review it periodically and state whether they wish to continue — which turns what sounds like a formality into a scheduled opportunity to reopen the terms. review in 2026, which is the single largest identifiable risk to the regional outlook and is treated in Part III.

10.3Colombia — the fiscal question arrives with the new government

Colombia grew 2.6 per cent in 2025 with inflation at 5.1 per cent, a general government deficit of 5.7 per cent of GDP and debt at 59.9 per cent.^{[IMF WEO]} The Economic Commission projects 2.7 per cent for 2026, noting that the recovery is consumption-led while investment has lagged and that activity has been driven "by market factors rather than by new economic policy stimulus."^{[CEPAL, Colombia country note 2025]}

Gustavo Petro's presidency ends in August 2026. His signature energy policy — no new hydrocarbon exploration licences — sits against an export book in which fuels and extractive products were 34.5 per cent of December 2025 export value, and crude output of 746,000 barrels a day in 2025 that has been broadly flat for a decade.^{[DANE via Agencia PI, see limits][EIA]} The 2025 export data show the adjustment already underway: petroleum and derivative exports fell 17.0 per cent and coal 31.0 per cent, while agricultural exports rose 33.2 per cent on a 70.2 per cent surge in unroasted coffee. Colombia is diversifying its export base, though partly by default rather than design.

Abelardo de la Espriella won the 21 June 2026 run-off with 49.66 per cent against Iván Cepeda's 48.70 per cent — a margin of under one percentage point, or roughly 252,000 votes on the registrar's nearly complete tally.^{[Reuters, 21–23 June 2026]} A sub-one-point margin in a country with an unresolved fiscal deficit, an active security deterioration and the region's largest resident Venezuelan migrant population is not a mandate for anything difficult.

10.4Chile — the region's institutional benchmark, tested

Chile grew 2.3 per cent in 2025 on IMF data with inflation at 4.2 per cent falling into the 3–4 per cent range by late 2025, a general government deficit of 2.8 per cent of GDP and debt at 41.8 per cent — the lowest of any large economy in the region.^{[IMF WEO][CEPAL, Chile country note 2025]} The Commission notes fiscal revenues came in below expectations during 2025, forcing spending restraint to hold the deficit near 2 per cent — which is precisely how a structural rule is supposed to behave. Copper's price strength and an appreciating peso worked in Chile's favour.

Chile is the region's mineral core: 5.3 million tonnes of mined copper in 2025 (23 per cent of world output), 180,000 kilotonnes of copper reserves (18.4 per cent of the world), 56,000 tonnes of lithium (the world's second-largest producer) and 9.2 million tonnes of lithium reserves.^{[USGS MCS 2026][USGS MCS 2026]} Ores and refined copper together were 54.7 per cent of 2024 goods exports.^{[UN Comtrade]} The state is now taking a direct position in lithium: Codelco and SQM completed the NovaAndino Litio joint venture in the Salar de Atacama, budgeting about US$3bn for direct lithium extractionPulling lithium straight out of brine with a chemical or membrane process, instead of evaporating the water in ponds over many months. It is faster and uses less land and water. It is also less proven at scale, which is where the capital risk sits. technology and targeting up to 470,000 tonnes of annual output against roughly 270,000 tonnes of current guidance.^{[Mining.com; Mining Weekly; Bloomberg]}

José Antonio Kast won the 14 December 2025 run-off with 58.17 per cent against Jeannette Jara's 41.83 on the Tribunal Calificador de Elecciones certification (Servel's preliminary count had been 58.16 / 41.84).^{[Servel / Tricel via Emol]} Chile has now changed governing coalition five times since 2010 — left to right in 2010, back in 2014, right again in 2018, left in 2022, right in 2026 — without ever suspending its fiscal rule, without a currency crisis, and without inflation above single digits for any sustained period. That is what a working political economy looks like. It is the benchmark for every other reform story in this report.

10.5Bolivia — the cycle, live, in phase III and IV simultaneously

If a reader wants to see the Dornbusch–Edwards sequence rather than read about it, watch Bolivia in 2026.

The setup was textbook. A gas boom from the mid-2000s funded a large expansion of the state and a fixed exchange rate against the dollar held since 2011. Gas production then declined without replacement reserves — Bolivian crude and condensate output fell from 51,000 barrels a day in 2014 to 18,000 in 2025 — while fuel subsidies continued.^[EIA] Reserves emptied. And the composition matters more than the headline: the Banco Central de Bolivia reports net international reserves of US$3,542.9m at 31 March 2026, down 4.6 per cent from end-2025 — but only US$144.4m of that was liquid foreign currency, a fall of 71.2 per cent in a single quarter, with 22.9 tonnes of gold worth US$3,432.2m accounting for roughly 95 per cent of gross reserves.^{[Banco Central de Bolivia, Q1 2026]} A country whose usable foreign exchange amounts to about two days of imports does not have a reserve buffer; it has a bullion vault. A dollar shortage has run since early 2023, producing a parallel exchange rate and fuel-distribution failures. The general government deficit reached 11.6 per cent of GDP in 2025, output contracted 1.2 per cent, and inflation reached 19.5 per cent.^{[IMF WEO April 2026]}

Rodrigo Paz won the October 2025 run-off with 54.5 per cent, ending twenty years of Movimiento al Socialismo rule; his Christian Democratic Party holds 49 of 130 deputies and 13 of 36 senators — the largest bloc, and short of a majority.^{[PBS/AP; El Comercio]} The adjustment followed almost immediately: fuel subsidies were removed, raising gasoline prices by about 87 per cent and diesel by about 162 per cent. The reaction was equally textbook — a general strike and as many as 70 roadblocks paralysing the country for 53 days between 2 May and 23 June 2026. In late July 2026 Bolivia and the IMF reached agreement on a programme worth about US$1.9bn, the country's first multi-year Fund arrangement since 2006 and materially below the US$2.5–2.8bn the government had sought. The reserve composition explains why the government could not wait: gold can be sold, but selling the last of it is a one-way door.^{[Reuters via WKZO; Americas Quarterly]}

Bolivia is not a repudiation of the left; the outgoing government's problem was arithmetic, not ideology. And the cost is being paid now: real wages are falling, in the phase-IV manner the 1989 paper predicted, and the beneficiaries of the boom are bearing it. Bolivia also holds 23 million tonnes of identified lithium resources — more than any country on earth — and produces essentially none, which is the most expensive institutional failure in the region measured in foregone income.

10.6Peru — extreme political instability, unusually stable macro

Peru is the region's best evidence that political chaos and macroeconomic discipline are separable. Peru grew 3.4 per cent in 2025 with inflation of 1.5 per cent, a general government deficit of 2.4 per cent of GDP and public debt of 30.2 per cent — among the strongest fiscal profiles in the hemisphere.^{[IMF WEO]} It is the world's second-largest copper producer at 2.7 million tonnes in 2025 with 85,000 kilotonnes of reserves.^{[USGS MCS 2026]}

Politically it has been in continuous crisis. Pedro Pablo Kuczynski resigned in 2018; Martín Vizcarra was removed in 2020; Manuel Merino lasted six days; Pedro Castillo attempted to dissolve Congress in December 2022 and was removed and arrested; Dina Boluarte was removed by a 122–0 congressional vote on 10 October 2025 on grounds of "permanent moral incapacity"; her successor José Jerí was himself removed on 17 February 2026.^{[CNN; Al Jazeera; Vatican News; Directorio Legislativo]} Keiko Fujimori won the 7 June 2026 run-off with 50.135 per cent against Roberto Sánchez's 49.865 per cent — a margin of 49,641 votes on the official ONPE tally declared on 3 July — and was inaugurated on 28 July 2026 as Peru's first elected female president and roughly the tenth head of state since 2016.^{[ONPE via Reuters, 3 July 2026; AP via NPR/Al Jazeera]}

Why does the macro survive? Because the fiscal and monetary regime is administered by a technocratic central bank and finance ministry whose personnel and rules have persisted across every one of those presidencies. Peru's institutions are weak where they are political and strong where they are technical. That is an unusual and, so far, durable configuration, and it is the strongest counter-example in the region to the claim that political instability necessarily produces macroeconomic instability.

10.7Ecuador — dollarised, adjusting, and losing the security battle

Ecuador grew 3.7 per cent in 2025 after contracting 2.0 per cent in 2024, with inflation of 0.7 per cent, a general government deficit of 2.9 per cent of GDP and debt at 54.4 per cent.^{[IMF WEO][CEPAL, Ecuador country note 2025]} Like El Salvador it is dollarised, so it cannot monetise a deficit; the Commission notes the overall fiscal result was near balance in 2025 but had retreated from a 2024 surplus, and that the oil sector remains persistently weak. Crude output has drifted from 556,000 barrels a day in 2014 to 441,000 in 2025.^[EIA]

The security picture is the opposite of El Salvador's, and the contrast is the most useful comparison in this report. Daniel Noboa declared an internal armed conflict in January 2024, designated criminal groups as terrorist organisations, deployed the military and captured several high-profile leaders. Ecuador's homicide rate then rose 31.2 per cent in 2025 to a record 50.9 per 100,000, the highest in the region outside Haiti.^{[InSight Crime 2025 Homicide Round-Up]} In November 2025 voters rejected all four of his referendum questions, including a constituent assembly and the return of foreign military bases, by roughly 61 per cent.^{[CNN Español; La Nación]}

The lesson is uncomfortable for anyone selling the Bukele model as a template. Ecuador applied militarisation, mass arrests and emergency powers, and violence rose. El Salvador applied a comparable toolkit and violence collapsed. The difference is structural: El Salvador's gangs were territorial extortion rackets with no external revenue, so incapacitating their members destroyed the business; Ecuador's are nodes in a transnational cocaine supply chain whose revenue comes from Colombian production and European demand, so removing personnel changes nothing about the profit pool. Mechanism Mass incarceration works against criminal organisations whose income is local and fails against those whose income is global. That single distinction predicts which countries can copy El Salvador and which cannot, and it is treated again in §13.

10.8Uruguay and Paraguay — the quiet cases that prove the argument

Uruguay grew 1.8 per cent in 2025 with inflation at 4.7 per cent, a deficit of 3.4 per cent of GDP and debt at 65.7 per cent; the Commission projects 2.1 per cent for 2026.^{[IMF WEO][CEPAL BP-2025]} Yamandú Orsi of the Frente Amplio took office in March 2025 after defeating the incumbent centre-right coalition, and the transition was uneventful — which is the point. Uruguay's economy is agricultural (soy, beef, dairy, cellulose) and is subject to the same terms-of-trade swings as everyone else's, plus drought. It has had no hyperinflation, no default since 2003, and no constitutional crisis. Its per-capita GDP of about US$24,500 is the highest in Latin America.^{[IMF WEO]} The mechanism is not mysterious: durable programmatic parties, an independent central bank, and a political culture in which the fiscal accounts are not a partisan question.

Paraguay grew 6.0 per cent in 2025 — the fastest in South America — with inflation at 4.1 per cent and public debt of 38.1 per cent of GDP.^{[IMF WEO]} The Commission records a deliberate consolidation from a 4.1 per cent deficit in 2023 to 2.6 per cent in 2024, a 1.9 per cent target for 2025 and 1.5 per cent for 2026, consistent with the upper bound of the Fiscal Responsibility Law, with public debt at US$19.92bn or 41.1 per cent of GDP in September 2025.^{[CEPAL, Paraguay country note 2025]} Santiago Peña, a former central banker and IMF staffer, is pursuing OECD accession. Paraguay's endowment is soy, beef and hydroelectricity — it co-owns Itaípú, one of the world's largest dams, with Brazil — and its comparative advantage is a very low tax burden. It is also the region's most dollar-exposed small economy in a different sense: its growth is a derivative of Brazilian and Argentine demand.

10.9Guyana and Suriname — the new petro-states, and the institution that will decide them

Guyana is the fastest-growing economy on earth and it is not close. Real GDP grew 43.5 per cent in 2020, 63.3 per cent in 2022, 43.8 per cent in 2024 and 19.3 per cent in 2025, with 16.2 per cent projected for 2026; nominal GDP has gone from US$4.79bn in 2018 to a projected US$33.96bn in 2026, a multiple of 7.1.^{[IMF WEO April 2026]} Stabroek blockThe offshore licence area holding Guyana's oil discoveries. A block is simply the mapped tract a government awards for exploration and production; in Guyana's case a single block is effectively the entire national industry. production passed 900,000 barrels a day in November 2025 after Yellowtail reached its 250,000-barrel design capacity; the co-venturers have committed more than US$60bn across seven sanctioned projects and expect 1.7 million barrels a day of capacity from eight developments by 2030.^{[ExxonMobil Guyana]} Irfaan Ali was re-elected on 1 September 2025 with over 55 per cent of the vote and 36 of 65 seats.^{[AS/COA; Al Jazeera; FTI]}

Everything about Guyana's future turns on layer 4 of the rent chain, and there is a real institution there to watch.

Guyana's Natural Resource Fund has received about US$9.3bn since 2020, of which US$7.8bn had been paid in by September 2025 on ExxonMobil's count, and held roughly US$3.96bn at end-May 2026 per the Bank of Guyana. The approved 2026 withdrawal is US$2.37bn — the first year the drawdown has fallen year on year, from US$2.46bn — financing about 31.7 per cent of a record GY$1.558trn (about US$7.47bn) national budget.^{[Bank of Guyana / Ministry of Finance via Stabroek News, Caribbean Council, INews Guyana]}

That is a fiscal rule doing exactly what a fiscal rule is for, and a 2026 withdrawal that fell rather than rose is the most encouraging thing about Guyana's institutional trajectory. It is also, on the projections, temporary: withdrawals are scheduled to climb to US$4.33bn by 2029. Inference Guyana's resource-curseThe long-observed pattern in which countries that strike a large natural resource end up growing more slowly, governing worse and borrowing more than comparable countries that never found anything. The curse is not the resource. It is what an unconstrained state does with income it did not have to earn. question will be settled in the 2027–2029 window, when the withdrawal path steepens and the political pressure to breach it will be at its maximum — not now, while the fund is still visibly accumulating.

Suriname is Guyana five years earlier. Its economy is small (US$4.66bn in 2025), its inflation high at 9.2 per cent, its general government deficit 9.6 per cent of GDP and its public debt 105.8 per cent — the highest ratio in the region.^{[IMF WEO]} Crude output is a residual 12,000 barrels a day. That changes at the end of the decade: TotalEnergies and APA Corporation took a final investment decisionThe moment a project's owners commit the capital irrevocably and construction begins. Everything before it is an option; everything after it is a schedule. In oil it is the only date worth putting in a diary. on the GranMorgu development in Block 58, a project of about US$10.5bn developing the Sapakara and Krabdagu discoveries with recoverable reserves estimated above 750 million barrels and a 220,000-barrel-a-day floating production unitA vessel moored over an offshore field that receives, processes and stores the oil until a tanker collects it. It stands in for the pipelines and onshore terminals a shallow-water field would need, which is how a country with no oil infrastructure at all can develop a deepwater discovery., with first oil targeted for 2028.^{[TotalEnergies; Staatsolie; APA Corporation]} A country with 105.8 per cent debt-to-GDP is about to receive a rent worth several multiples of its current GDP. Whether it builds a layer-4 institution before the money arrives is the only question that matters there, and it has roughly two years to do it.

§11Central America and the Caribbean: the remittance economies and the two failures

The isthmus runs on a different engine from South America. South American external earnings come from commodities sold into world markets. Central American external earnings come, to a first approximation, from labour exported to the United States and from services sold to United States consumers. Both are rents in the technical sense — income that does not require domestic productive capacity — but the second is far more sensitive to a single foreign government's policy.

The Economic Commission expects Central America to grow 3.0 per cent in 2026 after 2.6 per cent in 2025, describing the subregion as "the most affected by the weakening of external aggregate demand, particularly from the United States" and structurally dependent on it "in trade, finance and migration."^{[CEPAL BP-2025]}

11.1Guatemala — the largest economy, the weakest state capacity, the best recent news

Guatemala grew 4.1 per cent in 2025 with inflation of 1.6 per cent, a general government deficit of 1.9 per cent of GDP and public debt of 27.2 per cent — the lowest ratio in the region.^{[IMF WEO]} That low debt is not prudence so much as a tax take too small to finance a state: Guatemala's chronic problem is not fiscal excess but fiscal absence. The Commission records a central government deficit widening to 3.3 per cent in 2025 from 1.0 per cent in 2024 on higher public spending, with a current account surplus of 4.4 per cent of GDP driven by remittances.^{[CEPAL, Guatemala country note 2025]} Guatemala received US$25,530m of remittances in 2025, 53.5 per cent of the Northern Triangle total and roughly 21.4 per cent of its GDP.^{[IADB via Infobae]}

The political story is the region's most encouraging rule-of-law development. Bernardo Arévalo took office in January 2024 against sustained legal attempts to prevent it, and spent two years unable to remove Attorney General Consuelo Porras, who was protected by statute. In May 2026 he appointed a new attorney general, ending that struggle; 2026 also saw the naming of all five Constitutional Court members and the Electoral Tribunal.^{[Human Rights Watch, April 2026; Americas Quarterly; AP via WSB]} Guatemala is the one country in this report where the constraint on the executive has been strengthened rather than weakened during the period under review.

11.2Honduras and Nicaragua — two remittance economies, opposite institutional directions

Honduras grew 3.7 per cent in 2025 with inflation of 4.6 per cent, a general government deficit of just 0.7 per cent of GDP and debt of 41.2 per cent.^{[IMF WEO]} It captured US$12,212m of remittances in 2025, which the Inter-American Development Bank puts at 30.4 per cent of GDP — the highest dependence in Latin America.^{[IADB via Infobae]} Nasry Asfura of the National Party was declared president-elect for 2026–2030 by the National Electoral Council on 24 December 2025 with 1,479,748 votes, after an election conducted amid fraud accusations and boycott threats from several parties including from senior government figures.^{[CNE via HRN; Wikipedia summary of CNE declaration; EU EOM]} Honduras's homicide rate of about 31 per 100,000 in 2025 is the region's third-highest.^{[InSight Crime]}

Nicaragua grew 4.9 per cent in 2025 with inflation of 2.1 per cent, a general government surplus of 2.1 per cent of GDP and debt of 34.8 per cent — on paper, the soundest fiscal position in Central America.^{[IMF WEO]} Remittances are about 30 per cent of GDP. And it is the region's most complete authoritarian consolidation outside Cuba. The 2025 constitutional reform, passed by partial amendment through a regime-controlled legislature rather than by an elected constituent assembly, altered 148 of the constitution's 198 articles and eliminated 37 outright. Article 8 now vests state power in "the Presidency of the Republic, which directs the Government, and coordinates the Legislative, Judicial and Electoral organs"; Article 133 constitutes the presidency as a male and a female Co-President; and the presidential term was retroactively extended from five to six years.^{[ConstitutionNet; Verfassungsblog]} Daniel Ortega and Rosario Murillo, husband and wife, hold both seats.

Side by side, the two make §5 concrete: Nicaragua has better fiscal numbers than Honduras and a worse political system than anywhere in the hemisphere except Cuba. The fiscal anchor and the institutional constraint are genuinely independent, and a report that graded countries on the first alone would rank Nicaragua near the top of Central America.

11.3Costa Rica and Panama — the services economies

Costa Rica grew 4.6 per cent in 2025 with negative headline inflation of −0.1 per cent, a general government deficit of 3.4 per cent of GDP and debt of 60.4 per cent.^{[IMF WEO]} The Commission records a central government primary surplus of 1.3 per cent of GDP in 2025 against 1.1 per cent in 2024, with the financial deficit narrowing to 3.3 per cent from 3.8 per cent and public debt below 60 per cent of GDP for a second consecutive year.^{[CEPAL, Costa Rica country note 2025]} Costa Rica is the region's clearest case of a country that escaped commodity dependence entirely: medical devices, semiconductors, and high-value services. Laura Fernández won the 1 February 2026 election with 48.3 per cent, avoiding a run-off, and her Pueblo Soberano party took 31 legislative seats — the first absolute legislative majority since 1990.^{[TSE via La Nación / El Financiero]} That majority is worth watching precisely because Costa Rica's political system has been built on the absence of one.

Panama grew 4.4 per cent in 2025 with zero inflation, a deficit of 3.7 per cent of GDP and debt at 56.6 per cent.^{[IMF WEO]} Its rent is geographic. Panama Canal Authority revenue of US$5.7bn in fiscal 2025 generated a US$2.965bn transfer to the national treasury, up 20 per cent on the prior year, of which US$2.372bn was operating surplus and US$591m transit tonnage fees; canal revenue represents roughly a fifth to a quarter of Panamanian government revenue.^{[ACP via gCaptain; S&P Global]} Transits ran 10,726 in the October–June period of fiscal 2026, up 5.2 per cent, but the daily average of about 33 remains below pre-drought levels, and further draft restrictions are considered likely if the El Niño forecast materialises. Panama also holds the region's largest suspended asset: the Cobre Panamá copper mine, closed since 2023, on which President José Raúl Mulino committed to a decision by mid-2026 — permanent decommissioning or negotiated restart — with First Quantum having withdrawn its arbitration in March 2026.^{[Mining.com; Panama Times]} A Panamanian rent that depends on rainfall and a copper mine that depends on a court ruling is an unusually literal illustration of external dependence.

11.4The Dominican Republic, Belize, and the two failures

The Dominican Republic grew 2.1 per cent in 2025 on IMF data and 2.9 per cent on the Commission's, with inflation of 3.9 per cent, a deficit of 3.6 per cent of GDP and debt at 59.0 per cent.^{[IMF WEO][CEPAL, Dominican Republic country note 2025]} The Commission attributes the slowdown to weaker private investment and a decelerating tourism sector, and notes that tax collection was supported by higher mining revenues and capital gains. Luis Abinader's administration has run the region's most successful tourism-plus-mining-plus-remittances model, and the country's current account deficit narrowed to 2.7 per cent of GDP.

Belize is the region's smallest economy at US$3.33bn, growing 2.7 per cent in 2025 with inflation of 1.1 per cent and debt of 67.9 per cent of GDP.^{[IMF WEO]} Its significance to the regional story is genuinely marginal — the mandate asks that this be said explicitly rather than implied by omission — but it carries one instructive fact. The World Bank records that Belize's reform programme cut public debt from 103 to 62 per cent of GDP, with unemployment at a historic low of 2.1 per cent, in an economy of about US$3bn and 420,000 people where tourism contributes 45 per cent of GDP.^{[World Bank, May 2026]} Belize demonstrates that small size is not itself an obstacle to fiscal repair.

Haiti is the region's completed state failure and cannot be described in the same vocabulary as the rest. Real GDP contracted 2.7 per cent in 2025 after 4.2 per cent in 2024, with inflation at 28.3 per cent; the Commission projects a further 1.2 per cent contraction in 2026.^{[IMF WEO][CEPAL BP-2025]} The homicide rate reached about 68 per 100,000 in 2025 on 8,100 killings recorded through November by the United Nations Integrated Office in Haiti, and the UN warned that gang control of territory prevented verification in some areas, so the true figure is higher.^{[InSight Crime; BINUH via InSight Crime]} More than 2,000 people were killed in the first months of 2026 and nearly 1.5 million are internally displaced; armed groups control roughly 85 per cent of the Port-au-Prince metropolitan area and have expanded into at least five of Haiti's ten departments.^{[Haitian Times; International Crisis Group; NPR; OHCHR]} The Transitional Presidential Council's mandate expired on 7 February 2026. In September 2025 the Security Council converted the Multinational Security Support mission into a Gang Suppression Force with a UN support office; its leadership sought a two-year extension in July 2026.

Haiti matters to this primer for a reason beyond humanitarian urgency. It is the counterfactual that disciplines every other section: a country with no commodity rent at all, and therefore no macroeconomic populism in the Dornbusch–Edwards sense, has nonetheless failed more completely than any rent-state in the region. The rent is not the cause of state failure. The absence of a functioning state is, and the rent merely determines what form the failure takes.

§12The external actors: who actually holds what

The region's political economy is not decided only inside it. Five external actors hold instruments that materially change what a Latin American government can do, and several of those dependencies stay invisible until the actors are named.

That is a real change. For twenty years the standard framing of external influence in Latin America was a contest between American conditionalityThe policy commitments a lender attaches to its money: deficit targets, subsidy reform, a currency allowed to float. It is what makes a Fund loan different from a bond. It is also why a government sometimes wants the conditions as much as the cash — they let it do at a creditor's insistence what it could not do on its own authority. and Chinese capital, in which China's advantage was that its money came without political conditions. That framing is now out of date in both halves. Chinese policy-bank lending has collapsed from tens of billions a year to under a billion in 2022, and the lending that has resumed is small and project-tied.^{[Inter-American Dialogue / BU GDPC]} Meanwhile the United States has added a fourth instrument — direct military action to remove a head of state — to a toolkit that already included sanctions, swap linesA standing arrangement under which one central bank or treasury will lend another its currency on demand, in size, at short notice. It is the cheapest way to put dollars in a friendly country's hands in a crisis, and the fastest to take away. and the migration valve.

One consequence of that runs deeper than it looks. If Chinese influence now runs through demand rather than credit, it is much harder to sanction, much harder to reciprocate and much slower to change. A country can default on a Chinese loan; it cannot easily stop selling China its soy. Inference Washington's leverage over Latin American governments has grown in the short run and its leverage over Latin American economies has shrunk, because the thing that now binds the region to Beijing is a commercial relationship the United States cannot outbid at scale.

Part IIIThe frontier — durability, precedent, and the next cycle

§13Do the turnarounds last, and can either be exported?

Both turnarounds have now survived their first serious test. Neither has survived the only test that matters: a successor.

13.1Argentina past the referendum

The brief that commissioned this report expected the October 2026 midterms to be the referendum on Milei's adjustment. That referendum was held on 26 October 2025 and the government won it decisively — roughly 40.8 per cent of the national vote, sixteen provinces including Buenos Aires, and a haul reported at 64 of the 127 contested lower-house seats and 14 of the 24 contested Senate seats.^{[AS/COA; Americas Quarterly; NPR]} The stabilisation therefore no longer faces an electoral test until 2027. It faces three harder ones.

The reserve test. The end-2025 net international reserve target was missed. The programme requires accumulation of at least US$8bn during 2026, against roughly US$7.5bn of central bank purchases in the opening months.^{[IMF Art. IV 2026]} Argentina's stabilisations have never failed on the fiscal side; they have failed when the dollars ran out. The 2016–2018 episode had a plausible fiscal path and collapsed on external financing. This is the same risk in the same place, and it is the one thing on which the programme's own performance criteria have already slipped.

The federalism test. Provincial primary balances deteriorated by nearly a point of GDP in 2025 as borrowing constraints eased.^{[IMF Art. IV 2026]} A federal anchor in a federal country is a partial anchor. If the provinces continue to offset the centre, the general government position stops improving even as the federal accounts look pristine — which is precisely the pattern the 2024–2025 IMF and national numbers already show.

The succession test. La Libertad Avanza did not exist in 2020. The reform is currently a property of one man's political capital, not of a party organisation, a constitutional rule, or a cross-party consensus. Chile's fiscal rule survives alternation because it is law and because both coalitions accept it. Argentina has no equivalent. Inference The single most valuable thing the current government could do for the durability of its own programme is to convert the fiscal anchor into a statutory rule with automatic corrective mechanisms and provincial coverage — and there is no evidence in the Fund's 2026 documents that this is being attempted.

On the sugar-high question the verdict splits. The fiscal consolidation is structural: it came from spending cuts, subsidy reform and pension re-indexation, all of which are policy choices rather than price effects. The growth and the external improvement are substantially assisted by Vaca Muerta, where liquids output has risen past 1.04 million barrels a day.^[EIA] Both are real; they are not the same thing, and conflating them will produce a bad forecast in either direction.

13.2Bukele past the crackdown — and why the model does not travel

El Salvador's model faces a different problem: it has already delivered essentially all of its available benefit. Going from 105 homicides per 100,000 to 1.9 is a change that cannot be repeated, because there is almost nothing left to reduce. What continues to accrue is the cost — the state of exception renewed monthly for four years, a prison population at double capacity, several hundred custodial deaths, and a judiciary and legislature that no longer constrain the executive.

The exportability question is asked constantly and almost never answered with a mechanism. Here is the mechanism.

This matters well beyond Central America, because "the Bukele model" is now a live political export — invoked in Ecuador, Honduras, Costa Rica, Chile and Peru during the 2025–2026 election cycle. The mechanism above predicts that it will produce good results in countries whose criminal economies are domestic and extortive, and bad results in countries that sit on trafficking routes. Most of South America is the second kind.

13.3The Chilean counterfactual nobody campaigns on

One more comparison closes this section, because it is the region's most under-told story. Chile has, since 2006, run a centre-left coalition, then a right-wing government, then a centre-left coalition, then a right-wing government, then a left coalition, and now a right-wing government under José Antonio Kast — five changes of coalition in sixteen years. Across those six administrations inflation stayed in single digits with brief exceptions, public debt never exceeded roughly 42 per cent of GDP, the structural balance rule was never abolished, and the central bank was never used to finance the Treasury. Chilean per-capita GDP is US$17,735; Argentina's is US$14,355 and Venezuela's US$3,736.^{[IMF WEO April 2026]}

Nobody writes about Chile, because nothing happens there. That is the finding. The region's most successful political economy is the one that produces the least news, and the reforms that make headlines are almost always the ones being undertaken because something previously went badly wrong.

§14The Venezuela precedent: what the intervention makes possible, and fragile

Whatever else it was, the removal of a sitting Latin American head of state by United States military force in January 2026 was a change in the region's operating assumptions, and it is worth separating what is genuinely new from what is not.

What is not new: United States intervention in the hemisphere. What is new is the combination — direct military action against a state's leadership, followed within four weeks by a statutory reopening of that state's principal industry to United States companies, followed within eight weeks by a series of general licences authorising those companies to invest, followed within nine weeks by the restoration of diplomatic relations with the interim authorities the operation left in place. The sequence is unusually compressed and unusually commercial.

Three things follow from that sequence.

For the region's remaining authoritarian holdouts, the deterrent logic has changed in an ambiguous direction. The naive reading is that Havana and Managua are now more exposed. The evidence points the other way for Cuba, where the immediate consequence of Venezuela's change of management was not political pressure but the loss of its fuel supply — a shock that has produced grid collapses rather than liberalisation. Regimes under existential external pressure historically tighten rather than open. Inference The Venezuela precedent is more likely to produce accelerated repression in Nicaragua and Cuba than accelerated transition, at least in the near term, and the observable test is whether either state widens or narrows its space for private activity over the coming year. Cuba's July 2026 package of 176 measures, explicitly excluding privatisation of key sectors, is a data point in the tightening direction.

For China and Russia, the response has been notably thin. Russia condemned the operation and offered solidarity; it did not act.^[CSIS] China, which CSIS notes lent Venezuela more than US$60bn over recent decades, has an obvious commercial interest in the outcome of Venezuelan debt claims and no evident instrument for shaping it. The episode demonstrated that neither external patron could protect a client from direct American action, which is information every government in the region now possesses.

For sovereignty norms, the precedent cuts both ways and the honest position is uncertainty. The stabilising reading: an economy that had lost three-quarters of its output and driven 6.9 million people abroad now has a functioning oil sector, hundreds of political prisoners released, and a stated path to elections. The destabilising reading: a hemisphere in which the removal of a head of state by an external power is an available option has a materially different risk premiumThe extra return an investor demands for holding a riskier asset rather than a safe one. For a country it shows up as a higher borrowing cost, and it is simply the price the market puts on the chance that something goes wrong. on every sovereign asset in it, and every government now has a reason to hedge against Washington that it did not have in 2025. Both readings are supported. The report does not choose between them, because the evidence that would settle it — an electoral timetable, and what happens if it produces a government Washington dislikes — does not yet exist.

A transition is a schedule, not an intention. Until Caracas publishes one, everything else is atmosphere.

On the post-Maduro settlement

§15The next cycle: migration, minerals, and where the load will sit

15.1The migration reversal, and the remittance puzzle it explains

Two halves of the same 2025–2026 story have not been read together.

United States border encounters collapsed. October 2025, the first month of fiscal 2026, recorded 30,561 total nationwide encounters — the lowest start to a fiscal year ever recorded by Customs and Border Protection and 29 per cent below the previous record low of 43,010 in October 2012.^{[DHS, 5 November 2025]} By June 2026 nationwide encounters were 31,626 with 9,848 Border Patrol apprehensions on the south-west border, which CBP describes as 94 per cent below the prior administration's level and the lowest in over three decades.^{[CBP, 16 July 2026]}

Remittances to the Northern Triangle rose 20 per cent in the same year, to US$47,730.2m — El Salvador up 17.8 per cent to US$9,987.9m, Guatemala at US$25,530m, Honduras at US$12,212m.^{[IOM via EFE/Infobae; IADB]}

Those two facts look contradictory and are not. They are the same event, and the specialist literature says so directly: the Inter-American Dialogue's review of 2025 concludes that for every nationality except Mexico the growth "did not result from migration, but from migrants' risk-mitigation decisions in the event of deportation," against a backdrop of a 1 per cent United States excise on remittance cash transactions passed in the summer of 2025.^{[Inter-American Dialogue, 2 January 2026]} Mexico is the control: it was the only country to register a decline, because its migrant stock stopped growing and over 4 per cent of Mexican remitters stopped sending altogether.^{[Mexico Business News; Border Report]} The distinction that dissolves the paradox is between a stock and a flow. A flow is what migrants earn and send month by month; a stock is savings already accumulated, which can be moved in a single decision. Fear of deportation converts the second into the first, and a year of that makes the flow look enormous while emptying a reservoir that afterwards has to refill. Money was moved forward, not earned faster.

This is a falsifiable claim, and the first evidence is already in. If the 2025 surge was front-loading, 2026 growth must decelerate sharply. El Salvador's remittances rose 8.4 per cent in early 2026, less than half the 2025 rate, reaching US$2,435m in the first quarter.^{[BCR via Infobae, April 2026]} Mexico's fell 5.1 per cent over the first eleven months of 2025 to US$56.469bn before recovering 3.1 per cent in the first half of 2026 to US$30.759bn.^{[Banco de México via Mexico Business News]} Inference · testable If this reading is right, Northern Triangle remittance growth should continue decelerating through 2026–2027 towards or below the growth of the United States Hispanic wage bill, and the fiscal and external accounts of Honduras, El Salvador, Guatemala and Nicaragua — where remittances run from 21 to 30 per cent of GDP — will tighten accordingly. That is a slow-motion terms-of-trade shock to four countries at once, and none of them has a stabilisation fund for it.

15.2The energy transition and the region's actual leverage

The standard claim is that the energy transition hands Latin America enormous bargaining power through lithium and copper. The claim is half right, and the half that is wrong matters.

Copper is the strong hand. Chile and Peru produced 34.8 per cent of the world's mined copper in 2025 and hold 27.1 per cent of identified reserves; adding Mexico takes the region to 37.8 per cent of production and 32.4 per cent of reserves.^{[USGS MCS 2026]} Copper has no substitute at scale in electrification, the reserve base is not being replaced at the rate of depletion, and the COMEXThe New York exchange on which copper futures trade, and the reference price most physical contracts settle against. It is one of the markets §4 describes the region as a price-taker in. price averaged a projected record US$4.80 per pound in 2025. That advantage is durable, and it accrues to Chile and Peru specifically.

Lithium is the weak hand, and the USGS data show why. World lithium production rose 31 per cent in 2025 to about 290,000 tonnes excluding withheld United States output, while consumption rose 20 per cent to 263,000 tonnes; the average United States contract carbonateLithium carbonate: the processed chemical compound batteries actually consume, as distinct from the brine or ore it is refined from. The carbonate price, not the size of the resource, is what a lithium endowment is finally worth. price fell 31 per cent to US$9,000 per tonne.^{[USGS MCS 2026]} Lithium is abundant, its reserve base is being expanded faster than demand, and Latin America's share of production (Chile and Argentina together, 27.2 per cent) is far below its share of resources (42.7 per cent with Bolivia). A country holding resources in a market with excess supply holds an option, not an asset.

The distinction has a policy consequence that runs against the regional consensus. Chile's response — the state taking a direct position through the Codelco–SQM NovaAndino venture, with about US$3bn budgeted for direct lithium extraction and a target of up to 470,000 tonnes a year — is a bet on being a low-cost producer, which is the correct bet in an oversupplied market.^{[Mining Weekly; Bloomberg; Mining.com]} Bolivia's response — holding the largest identified resource and producing almost nothing while insisting on state control of the entire chain — is a bet that scarcity will arrive before the resource is stranded by chemistry. The price data do not currently support that bet.

15.3Where the load actually sits: five criticality tests

The mandate asks what happens when each load-bearing element moves. Here are the answers the evidence supports, stated as consequences rather than possibilities.

15.4What to watch

§16Five conclusions that run against the consensus

Each of these names the view it contradicts, rests on disclosed evidence, and can be shown wrong.

1. 1About two-fifths of Argentina's poverty collapse is arithmetic, not income. The consensus — on both sides of the Argentine debate — treats the fall from 52.9% to 28.2% as a pure test of the adjustment. The Fund's own decomposition attributes roughly 40% of the decline (7.3 percentage points) to the basic consumption basket rising 270% against headline inflation of 307%, mechanically moving households across a line defined by that basket. Labour income explains about 42%. Falsified if a later INDEC or Fund revision attributes materially less than a quarter of the decline to relative prices.
2. 2The Bukele model is a policy for extortion economies and will fail on trafficking routes. The consensus frames it as a trade between security and liberty that any country can choose to make. The mechanism says otherwise: mass incapacitation destroys a business whose revenue requires local personnel and does nothing to a business whose profit pool sits offshore. Ecuador ran a comparable playbook and its homicide rate rose 31.2% in 2025 to a record 50.9 per 100,000. Falsified if a transit-route country sustains a homicide decline of more than half under a mass-incarceration policy for three consecutive years.
3. 3The 2025 remittance record was a stock transfer, not a rise in earnings — so four Central American external accounts are already committed to a shock. Specialists have identified the cause: the Inter-American Dialogue's year-in-review states that for every nationality except Mexico the 2025 growth "did not result from migration, but from migrants' risk-mitigation decisions in the event of deportation." What is not being drawn is the consequence. Money brought forward is not money earned; the flow that arrives in 2027 is the underlying wage bill minus a depleting migrant stock, in four countries where remittances run from 21% to 30% of GDP and none of which has a stabilisation fund. Already partly confirmed: El Salvador's growth fell to 8.4% in early 2026 from 17.7% in 2025. Falsified if Northern Triangle remittance growth stays above 15% through 2026.
4. 4China's regional leverage got harder to counter precisely as its lending collapsed. The consensus treats retreating Chinese credit — new policy-bank commitments to the whole region were US$813m in 2022 against tens of billions a decade earlier — as a win for Washington. But influence that runs through demand rather than debt cannot be sanctioned, refinanced away or reciprocated; roughly 26–29% of Brazilian exports go to China. Falsified if a major regional exporter materially reduces its China trade share under United States pressure without a compensating loss of export value.
5. 5No Latin American authoritarian regime has fallen to economic collapse in twenty-five years. The consensus expects Cuban and Nicaraguan liberalisation to follow from economic distress. Venezuela lost 74.4% of real output between 2013 and 2020 and its regime survived until an external military operation removed it; Cuba has contracted for years and tightened; Nicaragua grew 4.9% in 2025 and became more repressive. Economic series are the wrong instrument for forecasting regime change. Falsified if either the Cuban or the Nicaraguan government undertakes a genuine transfer of power under internal economic pressure alone.

§17The strongest case against this report's thesis

What follows argues the opposing view at its strongest, not at its most convenient.

The structuralist case. The framework this report has used — macroeconomic populism, fiscal anchors, institutional constraint — is not a neutral description. It is a specific analytical tradition that locates the cause of Latin American underperformance inside Latin America, and in doing so it systematically understates the external determinants that a country cannot choose.

The evidence for that view is substantial. The region's terms of trade are set in Chicago, London and Shanghai. Its cost of capital is set by the Federal Reserve. Its largest economy's growth was cut to 0.6 per cent in 2025 in significant part by United States tariffs on its exports.^{[CEPAL, Mexico 2025]} Cuba's government estimates United States restrictions cost it US$7,556m between March 2024 and February 2025, equal to 79.4 per cent of its 2024 exports.^{[CEPAL, Cuba 2025, citing Cuban government figures]} Venezuela's collapse deepened under sanctions that removed its access to capital markets, to diluents and to its own refining assets abroad. And in January 2026 the head of state of a sovereign Latin American country was removed by a foreign military and replaced by an authority that promptly rewrote the country's oil law in favour of that foreign country's companies. On this reading, "institutional weakness" is a description of the symptom and the diagnosis is dependency: countries whose external earnings, capital costs and, in extremis, governments are determined abroad cannot build durable domestic institutions, and calling their failure to do so a policy error is blaming the patient.

This is not a fringe argument. It is the intellectual tradition that produced the Economic Commission for Latin America and the Caribbean itself, and a serious reader should be able to state it without caricature.

Where it fails. It fails on the cross-section, and it fails decisively. Every country in the region faces the same external environment. They face the same commodity prices, the same dollar cycle, the same Chinese demand, the same United States political weather. And their outcomes differ by an order of magnitude. Chile and Venezuela both had oil-and-metal endowments, both faced the 2014 price collapse, and one has per-capita income of US$17,735 while the other has US$3,736.^{[IMF WEO April 2026]} Peru has had nine presidents in a decade and public debt of 30.2 per cent of GDP; Brazil has had stable institutions and 93.3 per cent. Guyana went from nothing to the world's fastest-growing economy in six years under the same external conditions in which Venezuela, next door, with vastly larger reserves, could not maintain its own production. If external determinants explained outcomes, these dispersions could not exist.

The sanctions argument specifically fails on sequence. Venezuelan crude output had already fallen 41 per cent from its 2013 level by 2018, before the PDVSA sanctions took effect, and the collapse in upstream capital spending that produced that fall began years earlier.^[EIA] Sanctions made the trough deeper and the recovery slower. They did not start the descent.

What survives from the structuralist case, and should be conceded, is this: the external environment sets the amplitude and the timing of the cycle, and it determines the cost of a domestic mistake. A fiscal error in Chile is expensive; the same error in Suriname, with 105.8 per cent debt-to-GDP and a currency nobody holds, is existential. External conditions are not the cause. They are the multiplier — and for the smallest and most concentrated economies the multiplier is large enough that the distinction stops feeling academic. That is a real qualification and this report accepts it.

§18What would change this view

§19Limits of current understanding, and what is not verified

Three areas where this report's confidence is genuinely low, stated plainly rather than hedged into the prose.

Venezuela's macro data are not usable at the precision the tables imply. The IMF and the Economic Commission disagree about Venezuelan growth in 2025 by five percentage points (1.5 versus 6.5) and about inflation by more than an order of magnitude (252 per cent average on IMF staff estimates for 2025, against 23.6 per cent as of October 2024 on the Commission's reading of official series). Both are cited here because both are the best available from their respective institutions. Neither should be treated as a measurement.

The Venezuelan political settlement is unresolved and moving. Everything in §8 after 3 January 2026 describes a situation seven months old at the evidence cutoff, in a country where the governing authority has not published an electoral calendar. The report has framed that section around an open question rather than a verdict for exactly this reason.

Cuban data are official, partial and contested. Cuba is not an IMF member and does not appear in the World Economic Outlook. The 2025 figures used here are the Economic Commission's, derived from Cuban official series, and the 2026 figures are Cuban government statements reported by wire services. The two are not mutually consistent.

19.1Not verified — and carrying no conclusion

§20Evidence register

Tier 1 = issuer, regulator, multilateral or statistical-agency primary. Tier 2 = reputable wire, national outlet of record, or specialist institute. Tier 3 = single-source or aggregated. Every source below was opened and read during the research for this report on 4 August 2026. Event dates and publication dates are distinguished where they differ.

This is an internal explanatory primer built to deepen understanding of the region's political economy. It is not investment advice, not a recommendation, and not a solicitation. No security, issuer, sector or sovereign is graded, rated or recommended anywhere in this document. Evidence cutoff: 4 August 2026 — developments after that date are not reflected.