Milei and a Nobel Prize winner’s theory of decision-making in contexts of high uncertainty

Last Sunday, December 14th, I went to Olivos for a meeting with the President. We had several topics on the table, but one overshadowed the others for me. It was the topic that fascinated me the most.

Six days earlier, Javier Milei had received Robert “Bob” C. Merton (NE: Nobel Prize in Economic Sciences 1997) at the Quinta de Olivos. I wanted to understand what had happened in this conversation. Bob’s work has virtually shaped the way risk is assessed and decided under uncertainty. A conversation between them could influence the design of public policies, particularly in terms of their sequence and conditions of progress.

I asked the President what they talked about.

“Mainly real options and optimal decision”he replied.

I was silent for a moment.

“Oh really?”I managed to say.

The President noticed my surprise and laughed. He took paper and pencil and said to me:

“I’ll tell you what we discussed.”

What followed was not a business conversation. It was an attempt to formalize the problem. During these two hours, each reform appeared as a dynamic decision problem: possible actions, sunk costs, uncertainty and learning. Focus wasn’t the motto. It was the rule. When to commit. Which variable can be observed? What evidence is required? What threshold defines progress, correction, or abandonment?

If it is expensive to undo decisions, the cost is not just that they are wrong. It’s about making mistakes early on, coming to the conclusion with bad information and valuable options later. The problem is less about “choosing an action” and more about deciding when to exercise.

Merton is relevant because his ideas became infrastructure. His name has been associated with the Black-Scholes-Merton model. In 1997, he shared the Nobel Prize in Economics with Myron Scholes for a method for valuing derivatives that he had developed in collaboration with Fischer Black, who died before the award. This approach did not remain stuck in papers: it is implemented in models, controls and coverage used every day in banks, companies and regulators. The method is important: translate uncertainty into comparable rules, think of decisions as options and define criteria.

A choice is a right, not an obligation. You pay today to have the opportunity to vote tomorrow. If it’s right tomorrow, we’ll train. If tomorrow isn’t right, it will be let go. This asymmetry creates value.

Real options apply this logic to decisions in the real world, including government. In all cases, the problem is not just what to do, but also when and under what conditions to do it. Under uncertainty, when a decision is partially irreversible, waiting has value because it produces information. In many cases the criterion is not “appropriate or not appropriate” but rather “appropriate if a relevant variable exceeds a threshold”. Activity, inflation, conflict.

But waiting is hard. The rational decision compares at every point in time Marginal value of additional information compared to marginal cost of delay. It’s worth waiting until the costs outweigh the value of the information and flexibility gained.

The “optimal” decision comes from a practice rule: what is done today, what depends on results, what is checked, which indicator defines the threshold and which condition activates a correction or an exit. In politics, this structure allows for adjustment without the adjustment seeming like improvisation.

Let’s think about the decision to buy an apartment.

Buying today offers security but is placing a bet. Buying a home combines consumption and investment: a place to live and a large, illiquid asset with high transaction costs and slow payback. Irreversibility is real. Also uncertainty: employment, income, interest rate, credit, real estate cycle.

Waiting has a visible price: paying the rent. But maintain liquidity and buy information. It allows you to monitor rates and prices, stabilize income and set a horizon. Being an owner simultaneously pays a “dividend” in the form of stability: it reduces renegotiations and forced moves and grants control over the space.

Three characters organize the decision.

1. That the fee does not stifle income.

2. High probability of staying long enough to recoup entry costs and, if applicable, exit costs.

3. That the price is not out of the ordinary compared to comparable alternatives.

If these signs don’t appear, renting isn’t a waste: it pays for flexibility and keeps the decision open.

And there is a crucial warning: the option is not forever. The market is changing and the window is closing. Waiting only makes sense if you define what you are waiting for and what events change the decision.

They mostly occur in economic policy two symmetrical disturbances. Stiffnesswith closed packages designed as if there were no expectations and shocks. AND Volatilitywith corrections without rules that are perceived as improvisation.

In Olivos, Milei insisted on a third way: Flexibility with structure. The same goal and even the same measure can be better designed when formulated as a sequence with explicit conditions: What is being done now, what remains an option, what evidence is expected and what threshold triggers the next step.

This frame is ready to use with four parts.

Firstsequence reforms and define explicit assessments.

SecondInclude checks, triggers and processes. A correction occurs because an intended state is activated, not because the mood of the day has changed. For example, a rule that requires a review of a standard if an indicator does not improve within a defined period of time.

Thirdexpand private optionality by reducing friction and discretion.

RoomMake sure you have credibility as your core capital. If it increases, the number of possible decisions increases. If it falls, everything becomes more expensive and accelerates.

The fact that the president manages this framework has a concrete advantage: Convert goals into decision rules. The same program with the same content can be designed better if it includes processes, thresholds and reviews from the start. In addition, it allows something crucial: the introduction of explicit constraints on the branches of the decision tree, that is, determining in advance which paths will be automatically discarded. In the Argentine case, the only type of restrictions we will maintain from the start is simple and non-negotiable: that no trajectory violates ethical and moral values.

This is not aesthetics. It is the capacity of the government. Maintaining optionality in uncertain environments expands the number of viable policies, improves the balance between speed and risk control, and reduces remediation costs. It also improves coordination. None of this is a guarantee of success. But it increases the likelihood of getting it right, allowing you to learn without denying yourself, adapt without improvising, and move forward without losing degrees of freedom. The test will be simple and hard: follow the rules when the political costs will result in breaking them. In Argentina this difference is not semantic. It is the border between improvising and creating.