Back to writings
|Economics|8 min read

Why You Spend Pesos but Guard Dollars

At any given moment, the math is clean. One dollar equals a certain number of pesos. Both can buy the same thing at today's exchange rate. Static equivalence holds. And yet you will never see an Argentine reach for his dollar envelope when the almacΓ©n takes cash. He pays in pesos. Always.

This isn't superstition. It's a rational heuristic dressed up as a psychological bias β€” and it reveals something fundamental about how people actually think about money when they've been forced to learn its lessons the hard way.

Mental Accounting by Currency Tier

Richard Thaler won a Nobel Prize for noticing that people don't treat money as fungible. They put it in mental buckets β€” "fun money," "bill money," "savings" β€” and apply different rules to each bucket even when the amounts are identical. A $50 bonus feels different from $50 earned. A casino chip feels different from the same amount in cash.

Thaler called this mental accounting. He mostly wrote about it as a cognitive flaw, a deviation from the rational actor model. But there's a version of mental accounting that isn't a flaw at all.

Argentines sort currencies into tiers. Pesos sit in the "spend" bucket. Dollars sit in the "preserve" bucket. The psychological cost of parting with dollars is far higher β€” even at identical real value in the moment of transaction.

Bitcoin holders exhibit the same pattern. HODLers report visceral discomfort paying BTC for anything β€” coffee, concert tickets, a dinner bill. They'd pay $40 in fiat without blinking. They'll resist paying 0.00035 BTC for the same dinner, even knowing the exchange rate makes them equivalent today.

Static Equivalence vs. Dynamic Trajectory

Here's what economists miss when they call this a bias: the equivalence is static. The trajectory is not.

The Argentine peso has lost roughly 99% of its purchasing power over the past decade. The dollar loses maybe 3–5% per year in real terms β€” significant over decades, manageable over months. Bitcoin has compounded at rates that make any fiat comparison look embarrassing over any 4+ year window.

So when someone applies different mental accounting across currency tiers, they're not running an irrational bias. They're applying a forward-looking discount rate. They're asking: "What is this unit worth now versus what will it be worth when I need it most?"

The harder the money β€” the more predictable and resistant its supply is to debasement β€” the higher its expected future value relative to today. Spending hard money means giving up not just its current purchasing power but its compounding trajectory. The opportunity cost is asymmetric because the currencies are genuinely different goods with different properties. Equal today does not mean equivalent.

The Argentine as Proof of Concept

Argentina is the world's best natural experiment in monetary psychology. Citizens there don't have the luxury of abstract theory. They've watched currency after currency fail. They know, viscerally, what happens to soft money over time.

The result: a population that has spontaneously developed a two-tier monetary system without anyone instructing them to. Pesos are transaction tokens. Dollars are stores of value. The peso's role in the economy has been reduced to its minimum viable function β€” paying for things right now β€” because it cannot credibly perform any other monetary function.

This is Austrian economics not as theory but as lived reality. Mises and Hayek wrote about what happens when money loses credibility. Argentines live it every day. Their behavior is the prediction validated.

The peso spends. The dollar guards. The assignment is correct.

What This Reveals About Fiat Design

Here's where it gets uncomfortable.

If people naturally protect hard money and spend soft money, then an inflationary fiat system is β€” at least partially β€” engineered to remove that instinct. When every currency unit loses value over time, there's no hierarchy to sort into. Everything is soft. Everything belongs in the "spend" bucket. The psychological signal that would otherwise drive saving is deliberately neutralized.

Keynes understood this. He called the "euthanasia of the rentier" a feature, not a bug β€” a world where holding money is punished, where the only rational response to monetary debasement is to spend or speculate rather than save. The currency design makes frugality irrational.

Austrian economists call this the forced consumption effect of inflation: people are pushed up the time-preference curve against their will, consuming earlier than they would under sound money conditions.

What looks like normal consumer behavior is, in part, a rational response to a monetary environment that has removed the "guard" signal entirely.

Rational Heuristic or Cognitive Error?

So: is the hard money bias a cognitive error or a rational heuristic?

It's a rational heuristic with one genuine distortion.

The heuristic is sound: allocate spending from the currency most likely to depreciate, preserve the one most likely to hold or appreciate. This is correct. This is what every rational actor should do in a world of multiple currencies with different quality properties.

The distortion: people apply this asymmetry to opportunity cost calculations in ways that occasionally produce bad outcomes. The HODLer who won't buy a business tool for 0.001 BTC even when it would meaningfully improve his productive capacity is making an error β€” he's treating his sats as so sacred that he forgoes real present value. The Argentine who won't touch his dollar envelope even in genuine emergencies is overcorrecting.

The right move is to understand what the bias is actually tracking β€” quality-adjusted time preference β€” and apply it consciously rather than reflexively. Spend the soft money first, yes. But don't let the preservation instinct freeze you out of real decisions with real returns.

The framework is correct. The calibration requires judgment.

Conclusion

The Argentine with dollars under the mattress and the Bitcoin holder who won't pay sats for coffee are running the same mental model. They've sorted currencies by quality and applied asymmetric opportunity cost across tiers. In the essential sense, they are right.

The fact that this behavior reads as irrationality to most economists tells you something about the assumptions baked into mainstream models β€” specifically, the assumption that money is a neutral unit of account with no quality differential between units.

It isn't. It never was. And the people who've learned that lesson the hard way have built the correct mental framework for navigating a world of multiple competing currencies β€” even if they can't articulate the theory behind it.

The peso spends. The dollar guards. The sats stay cold.