Who Owns What Lies Beneath

There is film footage of Argentine railway stations in the interior. Platforms with grass growing through them. Station buildings in the style of the British railway companies that built them, windows smashed. Tracks that end in scrub after two hundred metres.

Around 1930 Argentina had one of the densest rail networks in the world — denser, relative to its area, than France’s. Much of it is now out of service. Many provincial capitals have no passenger connection at all. This is not a case of something never built. It is dismantling: a country that had an infrastructure and lost it again.

The obvious conclusion is that something went badly wrong here. The conclusion is correct. But it becomes more interesting once one asks when it went wrong. Because the answer lies not with the stations, but with a decision taken fifty years before they decayed, which had nothing to do with railways at all.

I. The First Decision

Around 1900 Argentina had the same starting conditions as the United States and Canada: temperate climate, exceptionally fertile soil, sparse settlement, mass immigration. By the logic of factor endowments the Pampas should have come to look like the American Midwest. Per head, the country was then as prosperous as France.

It did not look like that, and the reason was a political decision.

From 1862 the United States distributed its public land through the Homestead Act, in parcels, to settlers. After the campaign against the indigenous population — the Conquista del Desierto — Argentina distributed the Pampas in enormous blocks to a small number of families, partly as payment to those who had financed the campaign. Estancias of tens of thousands, sometimes hundreds of thousands, of hectares.

That set the pattern. Between 1880 and 1930 millions of immigrants arrived, more relative to population than in the United States. They could almost never become owners. They became tenants, farm labourers — or they went to Buenos Aires. Hence the extreme concentration of population in the capital, which persists today.

Why did this country not industrialise? The usual explanation is that the domestic market was too small. It does not quite convince, because Argentina was not poor.

The harder mechanism is political. The United States industrialised behind high protective tariffs, and Germany did the same from 1879. That requires a coalition that wants protection. In Argentina the elite had precisely the opposite interest: it exported beef and grain and imported everything else. Free trade was rational for them. There was simply nobody to push an industrial policy through.

And capital flowed where it found secure rent and standing — into land, not into factories. The railways, British-financed, were built radially toward the port of Buenos Aires, to carry exports out, not to connect a national market.

That is the key to the derelict stations. This network was never built for the country. It was built for the export trade. When the export model collapsed after 1930, half the lines lost their economic justification. A country’s infrastructure reveals what it was built for — and it decays when that purpose disappears.

II. Perón: First Consequence, Then Cause

Industrialisation, when it came, came involuntarily. The First World War interrupted imports. The Depression hit the most open country in the world hardest; exchange controls and the collapse of trade acted like a tariff nobody had legislated. The Second World War repeated the process.

Over those two decades an urban industrial working class formed in Buenos Aires, fed by internal migration from the interior. A large class for which the oligarchic system had made no provision, economic or political.

Perón did not invent this class. He integrated it. Labour law, real wage increases, paid holidays, pensions, health provision. He filled a gap the old system had left open for sixty years. Anyone who reads him merely as a demagogue cannot explain why the allegiance still holds.

The price lies in a single instrument. The IAPI, the state foreign trade agency, bought farmers’ harvests at fixed low prices and sold them at world market prices. The difference financed industry and social spending.

That was the lever — and the design fault. Perón taxed the one sector in which Argentina held a genuine competitive advantage in order to build up sectors in which it held none. Agriculture stopped investing, export earnings collapsed, and with them went precisely the foreign currency needed to import machinery for the industry.

This established a cycle Argentina has not left in seventy years: expansion, rising real wages, rising imports, trade deficit, devaluation, inflation, falling real wages, recession, and round again.

And a political structure that still stands: a permanent coalition of urban labour, protected industry and the state apparatus against the agricultural export sector. Each side uses the state against the other as soon as it holds power. Neither can eliminate the other.

The export taxes under dispute today are the direct descendants of the IAPI. The same grip on the same sector.

III. The First Rung and the Second

One might object: then Argentina simply has to catch up on industrialisation.

This is where we had to correct our own position. Because that route is closed.

Development economics knows the finding as premature deindustrialisation. Countries now reach their peak manufacturing share at considerably lower income per head than their predecessors did. Britain at its peak had nearly half its workforce in industry; Germany and Japan around a third. Many emerging economies peak at fifteen per cent — and then decline, long before prosperity has arrived.

The reasons: manufacturing has become more capital-intensive and absorbs fewer people per unit of output. Whoever enters today enters full global competition immediately — unlike Korea or Japan, which could build behind tariffs while supplying open Western markets. And the Chinese scale absorbs global demand growth without returning demand.

Counter-examples can be cited: Morocco, which became Africa’s largest vehicle exporter; Slovakia, which builds more cars per head than any country in the world — both squarely in China’s shadow. We examined them and no longer consider them sound. They are extended workbenches of European corporations. Their existence depends on the health of Renault, Stellantis and Volkswagen — and those are precisely the firms now under Chinese pressure. When the client shrinks, the workbench shrinks first. That is not independent industrialisation; it is an outsourced cost centre.

One case, however, withstands this diagnosis, and it is the most important. When the Multi Fibre Arrangement expired in 2005, the forecast was unanimous: China would absorb the textile market and Bangladesh would collapse. International programmes were prepared to cushion the expected mass layoffs. The opposite happened. In quota-free competition Bangladesh grew into the world’s second-largest garment exporter.

The entry point, then, has remained open. We hold to that.

What is more interesting is what happens twenty years later.

In August 2026 Bangladeshi industry has around 2,420 million cubic feet of gas per day against demand of 3,800 — a shortfall of thirty-six per cent. The Dhaka Chamber of Commerce puts the daily loss of output at up to 2,387 crore taka. In Gazipur roughly one factory in six has temporarily closed. Growth in industrial production fell to 2.86 per cent. And 1,857 investment applications worth some 35,000 crore taka are stalled because no new industrial gas connections are being issued.

At the same time, at an industry conference in Dhaka, the manufacturers’ association representative says the sector cannot remain competitive on cheap labour alone and must invest in productivity and skills. Named explicitly: the coming graduation from least developed country status — and with it the loss of tariff preferences, in particular duty-free access to the European market.

That describes the situation precisely. Bangladesh is not failing against China. It stands before the step onto the second rung and finds that this step costs exactly the three foundations the ascent rested on: low wages, tariff preferences, and low safety and environmental costs.

This is the uncomfortable point. The abuses documented since the Rana Plaza collapse of 2013 — a statutory minimum wage in the order of a hundred euros a month, enforced overtime, weak union rights, the dyeworks effluent in the rivers around Dhaka — are not accidents of this model. They are its condition. Remove them and you remove the competitive advantage. Leave them and you stay where you are.

The first rung can be climbed. The second costs you the first.

And Argentina stands worse: it cannot even reach the first. It was never cheap. It was rich.

IV. What Cannot Be Moved

The question therefore shifts. It is no longer how Argentina industrialises, but what it does with what it has.

And it has more than the images of decay suggest. Energy exports in 2023 averaged around 656 million dollars a month. In the first half of 2026 the figure was 1,270 million; May alone brought 1.78 billion. A doubling in three years. This is Vaca Muerta, one of the largest shale deposits in the world. Add lithium in the triangle with Chile and Bolivia, and copper in the cordillera, barely developed.

And there is capability nobody expects. Argentina leads the world in no-till farming; around that method grew an exporting industry for seed drills and precision agriculture. INVAP in Bariloche builds research reactors and has delivered them to Australia, Egypt and Algeria. Techint is an Argentine group of world rank in tubular products for the oil industry.

The pattern is strikingly consistent: it depends on something that cannot be relocated. A deposit, a soil, accumulated experience. China can build machine tools more cheaply. It cannot build Vaca Muerta.

This is precisely where the displacement mechanism does not reach.

V. Norway Is Not a Resource Story

At this point Norway is often invoked: a country that turned oil into prosperity. True — but true for a different reason than the one assumed.

When oil was found in 1969, parliament adopted the ten oil commandments in 1971. Their core: state control, a state oil company, an obligation on foreign corporations to involve Norwegian suppliers and Norwegian personnel, and revenues that accrue to society as a whole. Special taxation totals around seventy-eight per cent. The sovereign fund is the largest in the world; only the expected real return may be withdrawn.

Out of those requirements grew subsea technology, drilling technology, shipbuilding, engineering firms — a competence cluster that now sells worldwide, including where no Norwegian oil lies. A deposit was turned into transferable capability.

Norway is therefore not evidence that resources make you rich. Venezuela had the same oil. Norway is evidence that resources are not a curse only where it has first been settled who owns them and what happens to the proceeds. The Norwegian model is, at its core, a decision about ownership.

VI. The Third Decision

Argentina is making that decision now — and making it the other way round.

The RIGI, the special regime for large investments, applies from two hundred million dollars, with higher thresholds in oil and gas. It grants thirty years of tax, customs and foreign exchange stability, a fixed profit tax rate of twenty-five per cent, free access to foreign currency with the right to repatriate profits, and it opens the route to international arbitration.

Contrary to what is often claimed — and contrary to what we ourselves initially assumed — it does contain a requirement in favour of domestic value creation. An applicant must submit a plan for developing local suppliers and commit to paying at least twenty per cent of the investment sum to such suppliers. The obligation carries sanctions: breach can mean losing the benefits and repaying tax advantages with interest.

Argentina, then, did not capitulate unconditionally. And this shifts the comparison with Norway — it does not weaken it, it sharpens it.

Where do the two regimes actually differ? Not on local supply, but on three other points.

First, on capturing the rent. Norway taxes oil production at around seventy-eight per cent in total. Argentina levies a royalty of twelve per cent, which the provinces may under certain conditions reduce to five, plus twenty-five per cent on profits. The Ley Bases did not change that rate.

Second, on participation. Norway founded a state oil company and let the state earn from the licences. The RIGI provides for nothing of the kind.

Third, and this is what matters, on what the requirement is measured against. The twenty per cent is measured against the investment sum — that is, against construction. Norway’s requirements targeted ongoing operations and, explicitly, technology transfer. The difference is between a construction phase that generates contracts for civil works, logistics and catering, and a permanent relationship out of which drilling and subsea technology grow. Add to this that a local supplier is already anyone fiscally resident in the country or majority-owned domestically — a threshold a local subsidiary clears easily.

Put sharply: Argentina secured a share of the spending and gave away most of the return. Norway took both.

Since the constitutional reform of 1994 the mineral resources belong to the provinces. Whatever of the rent stays in the country therefore does not land where it could build structure either.

And here lies the point that cuts underneath the whole argument about state and market. The resource curse does not arise because a state takes too much. It arises because a rent flows in without productive effort and is then fought over.

A libertarian regime does not remove that mechanism. It relocates it. The rent then flows not to a political class but to concession holders. Who pockets it changes; that it is a rent does not.

In one respect it even becomes harder. Dutch disease — the overvalued currency that crushes everything else tradable — sets in regardless of who owns the production. Norway neutralised it by not letting the proceeds into the country in the first place. That instrument presupposes a state capable of acting.

And in Argentina the disease has already broken out before the resources are properly flowing. Since December 2023 the peso has fallen in nominal terms from 801 to around 1,500 to the dollar — a halving that would lead any observer to think of a weak currency. In real terms it has appreciated by some forty per cent over the same period. Because Argentine prices rose to more than three times their level, while the dollar rose only to 1.7 times.

That is why Argentina has become an expensive country measured in dollars — a finding readily attributed to regulation, which in fact sits in this difference.

The origin is what is remarkable. This appreciation does not come from the energy boom. It is a by-product of the stabilisation policy itself: use the exchange rate as an anchor against inflation and you appreciate in real terms. When Vaca Muerta delivers in full, the resource effect will therefore meet a currency that is already overvalued.

VII. What Speaks Against Us

We accept two objections.

The first: Argentina has no reserves and no access to capital markets. Norway could set conditions in 1971 because nobody doubted the country would remain solvent. Argentina has squandered that credit. After the emergency law of 2002, which ended the dollar peg and froze utility tariffs, came an avalanche of litigation; some sixty arbitration cases have been brought against the country in total, making it one of the most frequently sued states anywhere.

With that history behind you, predictability has to be bought by contract. In its stabilising part the RIGI is therefore not careless, but an attempt to substitute self-binding for missing credibility.

One may nonetheless ask whether the binding holds. The concessions of the 1990s also contained stability guarantees; in 2002 they were overwritten by statute. A contractual commitment does not bind a future legislature; it converts political risk into a later obligation to pay damages. What is sold is not security but a claim to compensation.

Above all, the concession covers only one part. Stability Argentina may have had to grant. Access to the rent — Norway’s seventy-eight per cent against twelve plus twenty-five, and a state participation that does not exist here — it did not therefore have to surrender. Norway did both at once in 1971, without any experience in the oil business, and the corporations came anyway. A host country’s bargaining power depends not only on its history but on the quality of what lies in the ground. Vaca Muerta counts among the best shale deposits in the world.

The one was necessity. The other was a choice.

The second objection weighs more and is aimed at our own proposed way out. The Norwegian path carries fewer people. It works for five million on an oil field. Argentina has forty-six million. Mining and energy employ few directly; a technology cluster employs engineers, not mass workforces.

So the real worry remains unanswered: there is at present no known model that lifts whole populations into prosperity. The path that carried many at once is closing. What remains open carries few. It is not a formulation we like, but we have not found a better one.

VIII. The Same Kind of Decision

In 1880 Argentina decided who owns the land, and it carries that decision still. In 1946 it decided who pays for industrialisation, and it has never left the cycle that followed. It is now deciding who owns what lies beneath.

Each time it is the same kind of decision, and each time it outlasts whoever made it.

What follows for us is closer to hand than it appears. We have shown elsewhere that German machine tool building is not being flooded but pushed out of its markets — and that half the loss does not come from China at all. There too the future lies not in becoming cheaper, but in what cannot be copied: thirty years of accumulated process knowledge.

Argentina could never build that knowledge, because it changed the rules every four years. Germany built it and has stopped paying attention.

Two ways of losing the same substance. The Argentine way is the more spectacular. The German way is the more avoidable.