WHERE AUSTRIAN ECONOMICS MEETS REAL WORLD ACTION

You have run the numbers twice.

The asset works. The price works. The financing works. You have tested it against a rate rise you do not expect and a slowdown you half do, and it still clears your hurdle.

And you are still not signing.

It is not the spreadsheet. You can price a bad rule — you have done it plenty of times. A tax you resent. A license regime you think is idiotic. A threshold that punishes you for hiring the eleventh person. You build it in, you demand a bigger return, you move on.

What stops you is not knowing which rule you will be under in five years.

Argentina has just spent five months writing a law about exactly that problem. The interesting part is not what the law says. It is what quietly came out of it along the way.

In March, Milei's government sent Congress a bill with a title you do not see often: the Inviolability of Private Property. It was drafted under Deregulation Minister Federico Sturzenegger and carries seven signatures — the President and six of his ministers.

Then it sat there for five months.

The text went through more than fifteen rewrites. Four separate times a vote was scheduled and did not happen, because the numbers were not there. It finally passed the Senate in the early hours of 7 August, 37 to 33, after nearly twelve hours of debate — and only after two whole chapters were cut to buy the votes. Rural land went two days before. Fire management went minutes before the vote.

What survived covers expropriation, evictions, title regularisation and the property registry. It is now sitting with the lower house, where the arithmetic is tighter still and the government has parked it while it repairs relations with provincial governors. Pass it unchanged and it becomes law. Change a word and it goes back to the Senate.

The law it rewrites has been on the books since 1977.

The government's own explanation of the bill does something you rarely see: it argues against its own record.

It points at the Heritage Foundation's economic freedom index and says that Argentina's overall score has just hit a 23-year high — and then, in the same breath, that the property rights component sits at 41 against a world average of 52. The high is the overall score, not property rights. On property rights specifically the government is telling you it is still well below everyone else.

It also admits a straight failure. By the government's own account, a programme meant to give informal residents legal title has been running for over seven years and has delivered title over 0.08 per cent of the land it was written for.

And Senator Patricia Bullrich told the chamber that the average wait to actually collect expropriation compensation in Argentina is eighteen years. That figure is hers and we have not been able to check it. Nothing in the existing law makes it hard to believe.

A newsletter with 'Austrian' in the title gets read as reflexively against whatever a government does. So here is the good news first, and there is a lot of it.

— They pay you before they take it. Title cannot pass until you have been paid in full. That one line closes the trapdoor behind an eighteen-year wait — take now, argue later.

— They cannot talk your price down first. Your compensation is based on what the asset was worth the day before the government said anything about wanting it. No announcing a compulsory purchase, watching the value collapse, then buying at the new price. The Senate actually improved on the original here.

— Inflation does not eat the cheque. The amount is indexed to inflation and carries interest. The original bill left the rate to whatever the judge thought commercially reasonable. The Senate replaced that with the Banco de la Nación's traditional fixed-term rate — a published number anyone can look up. Note which number they picked, though; we come back to it.

— You can claim lost trade, up to a point. If your business loses money because of the expropriation, that is now recoverable — capped at 30 per cent of the direct damage unless you can prove worse. The cap was the Senate's idea; the government had proposed no limit at all.

— Taking without taking now counts. If the state regulates your asset into uselessness without ever putting its name on the title, that can now be treated as an expropriation and compensated as one.

— Convenience is not an emergency. When the state wants to temporarily occupy your property, it now has to show real urgency, proportion, and that there was no gentler way of doing it. Administrative convenience is expressly not enough.

— Tenants are not just cleared out. Before anyone is removed where children, disabled people or vulnerable older people are living, the relevant protection bodies have to be brought in. Landlords must give ten days to fix rent arrears, not three. And a landlord can only get fast repossession on a sworn statement where his right is beyond argument — otherwise the judge can demand real security.

Any one of those would be a decent reform. Together they are the most serious attempt to make Argentine expropriation predictable since the law they are amending was written in 1977.

Take every word of that as true. What follows is unaffected.

The headline reform is a tighter test for when the state may take your property at all. It has to name the specific purpose. It has to be interpreted narrowly. And the taking has to be suitable, necessary and proportionate.

That is more than you get today, and it is worth having. But look at what kind of protection it is.

Today, Congress declares a public purpose and that is more or less that. Under the new law, a court can be asked afterwards whether the taking was proportionate. Which is real. It is also protection whose size is decided by somebody else, after the event, in a process you are not running.

The government's original draft had an anchor for this. It said the taking had to be suitable, necessary and proportionate — and then added, in the same sentence, that this applied where there was no less damaging way to achieve the same thing. That is a question with a findable answer. You point at the gentler option and ask why it was not used. The original also said the state could not single you out: the taking had to be non-discriminatory.

Both of those phrases came out in the Senate.

A lawyer will tell you the first one did not really go anywhere — that 'necessary' already means there was no gentler option, so deleting the explicit sentence changed nothing. He may be right. That is rather the point. What used to be written down is now something you would have to argue, in front of a judge, about a building you no longer have.

The non-discrimination line has no such fallback. It simply is not there any more.

You can price a rule. You cannot price a discretion.

And here is the tell. That deleted test — show there was no less damaging way — was written into the rules for temporarily occupying property at the same time it was taken out of the rules for permanently taking it. The drafters knew it was a good test. It survived where the stakes were lower.

Read the two halves of the bill side by side and something shows up that neither half shows on its own.

On what you get paid, the Senate moved towards hard numbers every single time. A published interest rate instead of a judge's. One consistent valuation date instead of two. A percentage ceiling on lost trade. Each of those swaps somebody's opinion for something you can look up.

On whether they can take it at all, it moved the other way every single time.

— The rule stopping the state from singling you out: gone.

— The rule requiring them to show there was no gentler option: gone.

— Regulated into uselessness: the original gave you two separate arguments — this restriction is so severe it violates my property right, and separately, this has effectively taken my property without the paperwork. The Senate merged them into one, and it is the harder one. A restriction only counts now if it also strips you of use.

— Temporary occupation: the original said 60 days, extendable once by 30, and carried a line saying that when the time was up the occupation ended by operation of law. The Senate made it 90 days, extendable once by another 90 — and deleted the line about it ending by itself.

So the state can move into your property for three months without expropriating it, and stay for six if it grants itself the extension. It does not buy it. It does not take title. It just occupies. You argue about the money afterwards — and nothing on the page now says the occupation stops on its own when the clock runs out.

Put plainly: the bill got more precise about what you are paid, and looser about when you can be taken from.

Those are not the same kind of thing and they do not cancel out. A precise payment formula only matters once someone has decided to take your building. The protections decide whether anyone decides that in the first place — and they are the half you have to price before you commit a penny.

The compensation fixes are real. They are also the smaller half. The question that stopped the deal at the top of this email was never 'what will they pay me if they take it'. It was 'how likely is this, and will I see it coming'.

Here is why that matters even if nothing ever happens to you.

Every buyer of Argentine property is asking the same question you were asking at the top of this newsletter. And every buyer answers it the same way — by paying less.

Which means the owner pays. Not eventually, not if the state ever comes. Today, in cash, at completion, as a discount on a sale to a private buyer for an expropriation that will almost certainly never happen to that building.

Multiply that by every transaction in the country. The cost of a rule nobody can price is being settled continuously, in private, by owners it never happens to. No notice is ever served. No compensation is ever calculated. The money moves anyway.

An expropriation regime is paid for mostly by people who are never expropriated.

That discount is the number this bill was written to shrink. It shrinks it a little — and then hands some of it back.

The interest rate is a small example of the same habit. Swapping a judge's view of a reasonable rate for a published bank rate is the right move: it is a rule where there was a discretion. But the rate they chose is what the bank pays depositors, not what it charges borrowers. In a country where those two numbers are a long way apart, an owner waiting years for his money is made whole for inflation and left short on the cost of the time. The rule is an improvement. The number inside it is cheap.

The law being rewritten here has stood since 1977. This bill amends it. A future Congress amends it back, on an ordinary Tuesday, with a simple majority.

The government proves the point itself, without meaning to.

Explaining why it wants to revisit the rural land rules, its own memorandum records that the relevant law was repealed by decree in December 2023, that a court has since suspended the effect of that repeal, and that as a result nobody is quite sure which version is currently in force.

That is Argentine property law, described by the ministry proposing to fix it.

Which is why the reversibility problem is not a debating point. Markets do not price the words in a statute. They price the rule they expect to be living under. And a pro-property law that squeaked through 37 to 33, lost two chapters on the way, and is now parked in the lower house while its sponsors go and find some votes, arrives with a reversal risk stapled to it.

The better the reform, the more someone will eventually want to undo it.

The Austrian Actor Issue #3 made this argument in the other direction. South African farmland repriced on the strength of a party resolution — no bill, no statute, nothing actually taken. Announcements move prices whether or not anything happens. That cuts both ways: a credible promise to protect property should lift asset values before a single article is enacted. Here it will barely register. Why it barely registers is the whole story.

One thing this bill is not: it is not a monetary measure in disguise. Nothing here touches the cost of borrowing, and there is no cheap-credit channel to trace. It works entirely through what people expect the rules to be. Different mechanism, and worth saying so rather than reaching for the nearest familiar explanation.

None of this is legal, tax or investment advice. It is analysis of a bill that is not yet law. Check anything you act on against the text and your own advisers.

Expropriation is a violation of property rights and no formula makes it otherwise. What follows assumes it is going to happen anyway, and asks what a law can still usefully deliver.

Not accuracy. Predictability.

The 30 per cent cap shows the right instinct and the wrong build. The instinct is correct: replace a judge working it out case by case with a published number. Three problems with this particular one:

— Nobody can say where 30 came from. It did not come from anywhere — it is a negotiated figure in a text that was rewritten fifteen times.

— It is a percentage of the direct damage, which is itself argued over. So it is one estimate expressed as a share of another estimate.

— You can exceed it by proving worse — which means every serious claimant will try, and you are back in front of a judge, now with 30 per cent sitting there anchoring the discussion.

Keep the instinct. Fix the build.

— Work the number out, do not pick it. Take what disruption actually cost owners in comparable past expropriations, as a percentage of what the property was worth, and average it. Percentages rather than pesos, because five years of Argentine inflation makes cash figures meaningless.

— Define "comparable" by something nobody can argue with. Group by transaction value in published bands. And fix the bands in the law or tie them to a published index — leaving them to be revised 'from time to time' quietly puts the discretion back in at the one place nobody looks.

— Only count what can be seen. Documented relocation costs. Contract penalties that were actually triggered. Time taken to get running again. Leave out lost future profit — not because it is not a real loss, but because nobody can measure a world that did not happen, and the alternative to leaving it out is making it up.

— Make collecting the data the price of the power. No expropriation proceeds unless what it cost is documented and filed to a public register. The state can take; the price of taking is recording what taking costs. If a legislature will not wear that, then where there is no data the figure must default to the highest observed — never to zero. Zero hands the benefit of the doubt to the party that controls both the power and the evidence.

Two honest limits. In year one the data is thin and the number is rough; it improves every year, and it is still better than an assessment, because the method is published and nobody gets to choose the answer.

And Argentina almost certainly does not have this data — because nobody has ever had to justify the number. That absence is the finding. Half a century of compensation law, and no record of what expropriation actually costs the people it happens to.

Sometimes the owner comes out ahead and sometimes he does not. That is the cost of the violation, not a fault in the formula.

Stay free,

Jean-Pierre

The Austrian Actor

P.S. The full archive will always be available at theaustrianactor.com