Friday, November 24, 2023

Pro Dollarization

With President Milei's election in Argentina, dollarization is suddenly on the table. I'm for it. Here's why. 

Why not? A standard of value

Start with "why not?'' Dollarization, not a national currency, is actually a sensible default. The dollar is the US standard of value. We measure length in feet, weight in pounds, and the value of goods in dollars. Why should different countries use different measures of value? Wouldn't it make sense to use a common standard of value? Once upon a time every country, and often every city, had its own weights and measures. That made trade difficult, so we eventually converged on international weights and measures. (Feet and pounds are actually a US anachronism since everyone else uses meters and kilograms.  Clearly if we had to start over we'd use SI units, as science and engineering already do.) 

Moreover, nobody thinks it's a good idea to periodically shorten the meter in order to stimulate the economy, say by making the sale of cloth more profitable. As soon as people figure out they need to buy more cloth to make the same jeans, the profit goes away. 


Precommitment is, I think,  the most powerful argument for dollarization (as for euorization of, say, Greece): A country that dollarizes cannot print money to spend more than it receives in taxes. A country that dollarizes must also borrow entirely in dollars, and must endure costly default rather than relatively less costly inflation if it doesn't want to repay debts. 

Ex post inflation and devaluation is always tempting, to pay deficits, to avoid paying debt, to transfer money from savers to borrowers, to advantage exporters, or to goose the economy ahead of elections. If a government can precommit itself to eschew inflation and devaluation, then it can borrow a lot more money on better terms, and its economy will be far better off in the long run. 

An independent central bank is often advocated for precommitment value. Well, locating the central bank 5,000 miles away in a country that doesn't care about your economy is as independent as you can get!

The Siren Vase. Greek 480-470 BC. Source: The Culture Critic

Precommitment is an old idea. See picture. It's hard. A country must set things up so that it cannot give in to temptation ex post, and it will regret and try to wriggle out of that commitment when the time comes. A lot of the structure of our laws and government amount to a set of precommitments. An independent central bank with a price-level mandate is a precommitment not to inflate. A constitution and property rights are precommitments not to expropriate electoral minorities. 

Especially in Argentina's case, precommitment is why full dollarization is better than an exchange rate peg or a currency board. A true exchange rate peg -- one dollar for one peso, as much as you like -- would seem to solve the temptation-to-inflate problem. But the country can always abrogate the peg, reinstitute currency controls, and inflate. An exchange rate peg is ultimately a fiscal promise; the country will raise enough taxes so that it can get the dollars necessary to back its currency. When that seems too hard, countries devalue the peg or abandon it altogether. 

A currency board is tougher. Under a currency board,  every peso issued by the government is backed by a dollar. That seems to ensure adequate reserves to handle any conceivable run. But a strapped government eyes the great Uncle-Scrooge swimming pool full of dollars at the currency board, and is tempted to abrogate the board, grab the assets and spend them. That's exactly how Argentina's currency board ended. Dollarization is a burn the ships strategy. There is no return. Reserves are neither necessary nor sufficient for an exchange rate peg. The peg is a fiscal promise and stands and falls with fiscal policy. 

A currency board, to the government

Full dollarization -- the country uses actual dollars, and abandons its currency -- cannot be so swiftly undone. The country would have to pass laws to reinstitute the peso, declare all dollar contracts to be Peso contracts, ban the use of dollars and try to confiscate them. Dollars pervading the country would make that hard. People who understand their wealth is being confiscated and replaced by monopoly money would make it harder -- harder than some technical change in the amount of backing at the central bank for the same peso notes and bank accounts underlying a devalued peg or even an abrogated currency board. 

The design of dollarization should make it harder to undo. The point is precommitment, to make it as costly as possible for a following government to de-dollarize, after all. It's hard to confiscate physical cash, but if domestic Argentine banks have dollar accounts and dollar assets, it is relatively easy to pronounce the accounts in pesos and grab the assets. It would be better if dollarization were accompanied by full financial, capital, and trade liberalization, including allowing foreign banks to operate freely and Argentinian banks to become subsidiaries of foreign banks. Absence of a central bank and domestic deposit insurance will make that even more desirable. Then Argentinian bank "accounts" could be claims to dollar assets held offshore, that remain intact no matter what a future Peronist government does. 

Governments in fiscal stress that print up money, like Argentina, also impose an array of economy-killing policies to try to prop up the value of their currency, so the money printing generates more revenue. They restrict imports with tariffs, quotas, and red tape; they can restrict exports to try to steer supply to home markets at lower prices; they restrict currency conversion and do so at manipulated rates; they restrict capital markets, stopping people from investing abroad or borrowing abroad; they force people to hold money in oligopolized bank accounts at artificially low interest rates. Dollarization is also a precommitment to avoid or at least reduce all these harmful policies, as generating a demand for a country's currency doesn't do any good to the government budget when there isn't a currency. 

Zimbabwe dollarized in 2009, giving up on its currency after the greatest hyperinflation ever seen. The argument for Argentina is similar. Ecuador dollarized successfully in much less trying circumstances. It's not a new idea, and unilateral dollarization is possible. In both cases there was a period in which both currencies circulated. (Sadly, Zimbabwe ended dollarization in 2019, with a re-introduction of the domestic currency and redenomination of dollar deposits at a very unfavorable exchange rate. It is possible to undo, and the security of dollar bank accounts in face of such appropriation is an important part of the dollarization precommitment.) 

The limits of precommitment

Dollarization is no panacea. It will work if it is accompanied by fiscal and microeconomic reform. It will be of limited value otherwise. I'll declare a motto: All successful inflation stabilizations have come from a combination of fiscal, monetary and microeconomic reform. 

Dollarization does not magically solve intractable budget deficits. Under dollarization, if the government cannot repay debt or borrow, it must default.  And Argentina has plenty of experience with sovereign default. Argentina already borrows abroad in dollars, because nobody abroad wants peso debt, and has repeatedly defaulted on dollar debt.  

The idea of dollar debt is that explicit default is more costly than inflation, so the country will work harder to repay debt. Bond purchasers, aware of the temptation to default, will put clauses in debt contracts that make default more costly still. For you to borrow, you have to give the bank the title to the house. Sovereign debt issued under foreign law, with rights to grab assets abroad works similarly. 

But sovereign default is not infinitely costly and countries like Argentina sometimes choose default anyway. Where inflation may represent simply hugging the mast and promising not to let go, default is a set of loose handcuffs that you can wriggle out of painfully. 

Countries are like corporations. Debt denominated in the country's own currency is like corporate equity (stock): If the government can't or won't pay it back the price can fall, via inflation and currency devaluation. Debt denominated in foreign currency is like debt: If the government can't or won't pay it back, it must default. (Most often, default is partial. You get back some of what is promised, or you are forced to convert maturing debt into new debt at a lower interest rate.) 

The standard ideas of corporate finance tell us who issues debt and who issues equity.  Small businesses, new businesses, businesses that don't have easily valuable assets,  businesses where it is too easy for the managers to hide cash, are forced to borrow, to issue debt. You have to borrow to start a restaurant. Businesses issue equity when they have good corporate governance, good accounting, and stockholders can be sure they're getting their share. 

These ideas apply to countries, and the choice between borrowing in their own currency and borrowing in foreign currency. Countries with poor governance, poor accounting, out of control fiscal policies, poor institutions for repayment, have to borrow in foreign currency if they are going to borrow at all, with intrusive conditions making default even more expensive. Issuing and borrowing in your own currency, with the option to inflate, is the privilege of countries with good institutions, and democracies where voters get really mad about inflation in particular. 

Of course, when things get really bad, the country can't borrow in either domestic or foreign currency. Then it prints money, forcing its citizens to take it. That's where Argentina is. In personal finance, you start with no credit at all; then you can borrow; finally you can issue equity. On the scale of healthier economies, dollarizing is the next step up for Argentina. 

Dollarization and foreign currency debt have another advantage. If a country inflates its way out of a fiscal mess, that benefits the government but also benefits all private borrowers at the expense of private savers. Private borrowing inherits the inflation premium of government borrowing, as the effective government default induces a widespread private default. Dollarization and sovereign default can allow the sovereign to default without messing up private contracts, and all prices and wages in the economy. It is possible for sovereigns to pay higher interest rates than good companies, and the sovereign to be more likely to default than those companies. It doesn't always happen, because sovereigns about to default usually grab all the wealth they can find on the way down, but the separation of sovereign default from inflationary chaos is also an advantage. 

Greece is a good example, and a bit Italy as well, both in the advantages and the cautionary tale about the limitations of dollarization. Greece and Italy used to have their own currencies. They also had borders, trade controls, and capital controls. They had regular inflation and devaluation. Every day seemed to be another "crisis" demanding another "just this once" splurge. As a result, they paid quite high interest rates to borrow, since savvy bondholders wanted insurance against another "just this once."

They joined the EU and the eurozone. This step precommitted them to free trade, relatively free capital markets, and no national currency.  Sovereign default was possible, but regarded as very costly. Having banks stuffed with sovereign debt made it more costly.  Leaving the euro was possible, but even more costly. Deliberately having no plan to do so made it more costly still. The ropes tying hands to the mast were pretty strong. 

The result: borrowing costs plummeted. Governments, people and businesses were able to borrow at unheard of low rates. And they did so, with aplomb. The borrowing could have financed public and private investment to take advantage of the new business opportunities the EU allowed. Sadly it did not. Greece soon experienced the higher ex-post costs of default that the precommitment imposed. Dollarizaton -- euroization -- is a precommitment, not a panacea. Recommitments impose costs on yourself ex post. Those costs are real.  

A successful dollarization for Argentina has to be part of a joint monetary, fiscal, and microeconomic reform. (Did I say that already? :) ) If public finances aren't sorted out, a default will come eventually. And public finances don't need a sharp bout of "austerity" to please the IMF. They need decades of small primary surpluses, tax revenues slightly higher than spending, to credibly pay down any debt. To get decades of revenue, the best answer is growth. Tax revenue equals tax rate times income. More income is a lot easier than higher tax rate, which at least partially lowers income. Greece and Italy did not accomplish the microeconomic reform part. 

Fortunately, for Argentina, microeconomic reform is low-hanging fruit, especially for a Libertarian president. 


Well, so much for the Promised Land, they may have asked of Moses, how do we get there? And let's not spend 40 years wandering the Sinai on the way. 

Transition isn't necessarily hard. On 1 January 1999, Italy switched from Lira to Euro. Every price changed overnight, every bank account redenominated, every contract reinterpreted, all instantly and seamlessly. People turned in Lira banknotes for Euro banknotes. The biggest complaint is that stores might have rounded up converted prices. If only Argentina could have such problems. 

Why is Argentina not the same? 

Well, for a lot of reasons. Before getting to the euro, Italy had adopted the EU open market. Exchange rates had been successfully pegged at the conversion rate, and no funny business about multiple rates. The ECB (really the Italian central bank) could simply print up euros to hand out in exchange for lira. The assets of the Italian central bank and other national central banks were also redenominated in euro, so printing up euros to soak up national currencies was not inflationary -- assets still equal liabilities. Banks with lira deposits that convert to Euro also have lira assets that convert to euro. And there was no sovereign debt crisis, bank crisis,  or big inflation going on. Italian government debt was trading freely on an open market. Italy would spend and receive taxes in euros, so if the debt was worth its current price in lira as the present value of surpluses,  it was worth exactly the same price, at the conversion rate, in euro. 

None of this is true in Argentina. The central problem, of course, is that the government is broke. The government does not have dollars to exchange for Pesos. Normally, this would not be a problem. Reserves don't matter, the fiscal capacity to get reserves matters. The government could simply borrow dollars internationally, give the dollars out in exchange for pesos, and slowly pay off the resulting debt. If Argentina redenominated interest-bearing peso debt to dollars at a market exchange rate, that would have no effect on the value of the debt. 

Obviously, borrowing additional dollars would likely be difficult for Argentina right now. To the extent that its remaining debt is a claim to future inflationary seigniorage revenues, its debt is also worth less once converted to dollars, even at a free market rate, because without seigniorage or fiscal reforms, budget deficits will increase. 

And that leads to the primary argument against dollarization I hear these days. Yes it might be the promised land, but it's too hard to get there. 

I don't hear loudly enough, though, what is the alternative? One more muddle of currency boards, central bank rules, promises to the IMF and so forth? How do you suddenly create the kind of stable institutions that Argentina has lacked for a century to justify a respectable currency? 

One might say this is a problem of price, not of quantity. Pick the right exchange rate, and conversion is possible. But that is not even clearly true. If the state is truly broke, if pesos are only worth anything because of the legal restrictions forcing people to hold them, then pesos and peso debt are genuinely worthless. The only route to dollarization would be essentially a complete collapse of the currency and debt. They are worth nothing. We start over. You can use dollars, but you'll have to export something to the US -- either goods or capital, i.e. stock and bonds in private companies -- to get them. (Well, to get any more of them. Lots of dollars line Argentine mattresses already.) That is enough economic chaos to really put people off. 

In reality, I think the fear is not a completely worthless currency, but that a move to quick dollarization would make peso and peso claims worth very little, and people would rebel against seeing their money holdings and bank accounts even more suddenly worthless than they are now. Maybe, maybe not. Just who is left in Argentina counting on a robust value of pesos? 

But the state is not worth nothing. It may be worth little in mark to market, or current dollar borrowing capacity. But a reformed, growing Argentina, with tax, spending, and microeconomic reform, could be a great place for investment, and for tax revenue above costs. Once international lenders are convinced those reform efforts are locked in, and Argentina will grow to anything like its amazing potential, they'll be stumbling over themselves to lend. 

So a better dollarization plan redeems pesos at the new greater value of the post-reform Argentine state. The question is a bit of chicken and egg: Dollarization has to be part of the reform, but only reform allows  dollarization with a decent value of peso exchange.  So there is a genuine question of sequencing of reforms. 

This question reminds me of the totally fruitless discussion when the Soviet Union broke up. American economists amused themselves with clever optimal sequencing of liberalization schemes. But if competent benevolent dictators (sorry, "policy-makers") were running the show, the Soviet Union wouldn't have failed in the first place. 

The end of hyperinflation in Germany. Price level 1919-1924. Note left-axis scale. Source: Sargent (1982) "The ends of four big inflations." 

A better historical analogy is, I think, the ends of hyperinflation after WWI, so beautifully described by Tom Sargent in 1982. The inflations were stopped by a sudden, simultaneous, fiscal, monetary, and (to some extent) microeconomic reform. The fiscal problem was solved by renegotiating reparations under the Versailles treaty, along with  severe cuts in domestic spending, for example firing a lot of government and (nationalized) railroad workers. There were monetary reforms, including an independent central bank forbidden to buy government debt. There were some microeconomic reforms as well. Stopping inflation took no monetary stringency or high interest rates: Interest rates fell, and the governments printed more money, as real money demand increased. There was no Phillips curve of high unemployment. Employment and the economies boomed. 

So I'm for almost-simultaneous and fast reforms. 

1) Allow the use of dollars everywhere. Dollars and pesos can coexist. Yes, this will put downward pressure on the value of the peso, but that might be crucial to maintain interest in the other reforms, which will raise the value of the peso. 

2) Instant unilateral free trade and capital opening. Argentina will have to export goods and capital to get dollars. Get out of the way. Freeing imports will lower their prices and make the economy more efficient. Capital will only come in, which it should do quickly, if it knows it can get out again. Float the peso. 

3) Long list of growth - oriented microeconomic reforms. That's why you elected a Libertarian president.  

4) Slash spending. Reform taxes. Low marginal rates, broad base. Subsidies in particular distort prices to transfer income. Eliminate. 

5)  Once reforms are in place, and Argentina has some borrowing capacity, redenominate debt to dollars, and borrow additional dollars to exchange pesos for dollars. All existing peso contracts including bank accounts change on the date. 

Basically, you want people to hold peso bills and peso debt in the interim as claims on the post-reform government. Peso holders have an incentive to push for reforms that will raise the eventual exchange value of the peso.   

6) Find an interim lender. The central problem is who will lend to Argentina in mid stream in order to retire pesos. This is like debtor in possession financing but for a bankrupt country. 

This could be a job for the IMF. The IMF could lend Argentina dollars for the purpose of retiring pesos. One couldn't ask for much better "conditionality" than a robust Libertarian pro-growth program. Having the IMF along for the ride might also help to commit Argentina to the program. (The IMF can force conditionality better than private lenders.) When things have settled down, Argentina should be able to borrow dollars privately to pay back the IMF. The IMF might charge a decent interest rate to encourage that. 

How much borrowing is needed? Less than you think. Interest-paying debt can simply be redenominated in dollars once you pick a rate. That might be hard to pay off, but that's a problem for later. So Argentina really only needs to borrow enough dollars to retire cash pesos. I can't find numbers, but hyper inflationary countries typically don't have much real value of cash outstanding. The US has 8% of GDP in currency outstanding. If Argentina has half that, then it needs to borrow only 4% of GDP in dollars to buy back all its currency. That's not a lot. If the peso really collapses, borrowing a little bit more (against great future growth of the reform program) to give everyone $100, the sort of fresh start that Germany did after WWII and after unification, is worth considering. 

Most of the worry about Argentina's borrowing ability envisions continued primary deficits with slow fiscal adjustment. Make the fiscal adjustment tomorrow.

"You never want a serious crisis to go to waste," said Rahm Emanuel wisely. "Sequencing" reforms means that everything promised tomorrow is up for constant renegotiation. Especially when parts of the reform depend on other parts, I'm for doing it all as fast as possible, and then adding refinements later if need be. Roosevelt had his famous 100 days, not a 8 year sequenced program. 

The Argentine reform program is going to hurt a lot of people, or at least recognize losses that had long been papered over in the hope they would go away. Politically, one wants  to make the case "We're all in this, we're all hurting. You give up your special deal, preferential exchange rate, special subsidy or whatever, but so will everyone else. Hang with me to make sure they don't get theirs, and in a year we'll all be better off." If reforms are in a long sequence, which means long renegotiation, it's much harder to get buy in from people who are hurt earlier on that the ones who come later will also do their part.  

The standard answers

One standard critique of dollarization is  monetary policy and "optimal currency areas." By having a national currency, the country's wise central bankers can artfully inflate and devalue the currency on occasion to adapt to negative shocks, without the inconvenience and potential dislocation of everyone in the country lowering prices and wages. 

Suppose, say, the country produces beef, and exports it in order to import cars. If world demand for beef declines, the dollar price of beef declines. The country is going to have to import fewer cars. In a dollarized country, or with a pegged exchange rate,  the internal price of beef and wages go down. With its own country and a floating rate, the value of the currency could go down, leaving beef and wages the same inside the country, but the price of imported cars goes up.  If lowering prices and wages causes more recession and dislocation than raising import prices, then the artful devaluation is the better idea.  (To think about this question more carefully you need traded and non-traded goods; beef, cars, and haircuts. The relative price of beef, cars, and haircuts along with demand for haircuts is also different under the two regimes). 

Similarly, suppose there is a "lack of demand'' recession and deflation. (90 years later, economists are still struggling to say exactly where that comes from.) With its own central bank and currency, the country can artfully inflate just enough to offset the recession. A country that dollarizes also has to import not-always-optimal US inflation. Switzerland did a lot better than the US and EU once again in the covid era. 

This line of thinking answers the question, "OK, if Argentina ($847 bn GDP, beef exports) should have its own currency in order to artfully offset shocks, why shouldn't Colorado ($484 bn GDP, beef exports)?''  Colorado is  more dependent on trade with the rest of the US than is Argentina. But, the story goes, people can more easily move across states. A common federal government shoves "fiscal stimulus" to states in trouble. Most of all, "lack of demand" recessions seem to be national, in part because of the high integration of states, so recessions are fought by national policy and don't need state-specific monetary stimulus. 

This is the standard "optimal currency area" line of thinking, which recommends a common currency in an integrated free trade zone such as US, small Latin American countries that trade a lot with the US, and Europe. Standard thinking especially likes a common currency in a fiscal union.  Some commenters felt Greece should keep or revert to the Drachma because the EU didn't have enough common countercyclical fiscal policy. It likes independent currencies elsewhere.

I hope you're laughing out loud by now. A wise central bank, coupled with a thrifty national government, that artfully inflates and devalues just enough to technocratically exploit price stickiness and financial frictions, offsetting national "shocks" with minimum disruption, is a laughable description of Argentina's fiscal and monetary policies. Periodic inflation, hyperinflation and default, together with a wildly overregulated economy with far too much capital and trade controls is more like it. 

The lure of technocratic stabilization policy in the face of Argentina's fiscal and monetary chaos is like fantasizing whether you want the tan or black leather on your new Porsche while you're on the bus to Carmax to see if you can afford a 10-year old Toyota. 

Another reason people argue that even small countries should have their own currencies is to keep the seigniorage. Actual cash pays no interest. Thus, a government that issues cash earns the interest spread between government bonds and interest. Equivalently, if demand for cash is proportional to GDP, then as GDP grows, say 2% per year, then the government can let cash grow 2% per year as well, i.e. it can print up that much cash and spend it. 

But this sort of seigniorage is small for modern economies that don't have inflation. Without inflation, a well run economy might pay 2% for its debt, so save 2% by issuing currency. 2% interest times cash which is 10% of GDP is 0.2% of GDP. On the scale of Argentinian (or US) debt and deficits, that's couch change.  

When inflation is higher, interest rates are higher, and seigniorage or the "inflation tax" is higher. Argentina is living off that now. But the point is not to inflate forever and to forswear bigger inflation taxes. 

Keeping this small seigniorage is one reason for countries to keep their currency and peg to the dollar or run a currency board. The currency board holds interest-bearing dollar assets, and the government gets the interest. Nice. But as I judge above, the extra precommitment value of total dollarization is worth the small lost seigniorage.  Facing Argentina's crisis, plus its catastrophic century of lost growth, lost seigniorage is a cost that I judge far below the benefit. 

Other countries dollarize, but agree with the US Fed to rebate them some money for the seigniorage. Indeed, if Argentina dollarizes and holds 10% of its GDP in non-interest-bearing US dollars, that's a nice little present to the US. A dollarization agreement with Argentina to give them back the seignorage would be the least we can do. But I don't think Argentina should hold off waiting for Jay Powell to answer the phone. The Fed has other fires to put out. If Argentina unilaterally dollarizes, they can work this sort of thing out later. 

Dollarization would obviously be a lot easier if it is worked out together with the US government and US banks. Getting cash sent to Argentina, getting banks to have easy payment systems in dollars and links to US banks would make it all easier. If Argentina gets rid of its central bank it still needs a payment system to settle claims in dollars. Accounts at, say, Chase could function as a central bank. But it would all be easier if the US cooperates. 


Some commenters point out that Argentina may be importing US monetary policy just as the US imports Argentine fiscal policy. That would lead to importing a big inflation. They suggest a Latin American Monetary Union, like the euro, or using a third country's currency. The Swiss franc is pretty good. Maybe the Swiss can set the world standard of value. 

Both are good theoretical ideas but a lot harder to achieve in the short run. Dollarization will be hard enough. Argentines have a lot of dollars already, most trade is invoiced in dollars so getting dollars via trade is relatively easy, the Swiss have not built out a banking infrastructure capable of being a global currency. The EMU lives on top of the EU, and has its own fiscal/monetary problems. Building a new currency before solving Argentina's problems sounds like a long road. The question asked was dollarization, so I stuck to that for now. 

I imagined here unilateral dollarization. But I didn't emphasize enough: The US should encourage dollarization! China has figured this out and desperately wants anyone to use its currency. Why should we not want more people to use our currency? Not just for the seigniorage revenue, but for the ease of trade and international linkages it promotes. The Treasury and Fed should have a "how to dollarize your economy" package ready to go for anyone who wants it. Full integration is not trivial, including access to currency, getting bank access to the Fed's clearing systems, instituting cyber and money laundering protocols, and so forth. 

Important update: 

Daniel Raisbeck and Gabriela Calderon de Burgos at CATO have a lovely essay on Argentinian dollarization, also debunking an earlier Economist article that proclaimed it impossible. They include facts and comparison with other dollarization experiences, not just theory as I did. (Thanks to the correspondent who pointed me to the essay.) 

Some quotes:

At the end of 2022, Argentines held over $246 billion in foreign bank accounts, safe deposit boxes, and mostly undeclared cash, according to Argentina’s National Institute of Statistics and Census. This amounts to over 50 percent of Argentina’s GDP in current dollars for 2021 ($487 billion). Hence, the dollar scarcity pertains only to the Argentine state....

The last two dollarization processes in Latin American countries prove that “purchasing” the entire monetary base with U.S. dollars from one moment to the next is not only impractical, but it is also unnecessary. 

In both Ecuador and El Salvador, which dollarized in 2000 and 2001 respectively, dollarization involved parallel processes. In both countries, the most straightforward process was the dollarization of all existing deposits, which can be converted into dollars at the determined exchange rate instantly.

in both Ecuador and El Salvador, dollarization not only did not lead to bank runs; it led to a rapid and sharp increase in deposits, even amid economic and political turmoil in Ecuador’s case....

There is a general feature of ending hyperinflation: People hold more money. In this case, people hold more bank accounts once they know those accounts are safe. 

Short summary of the rest, all those dollar deposits (out of mattresses into the banking system) allowed the central bank to retire its local currency liabilities. 

Emilio Ocampo, the Argentine economist whom Milei has put in charge of plans for Argentina’s dollarization should he win the presidency, summarizes Ecuador’s experience thus:

People exchanged their dollars through the banks and a large part of those dollars were deposited in the same banks. The central bank had virtually no need to disburse reserves. This was not by design but was a spontaneous result.

In El Salvador also, 

Dollar deposits also increased spontaneously in El Salvador, a country that dollarized in 2001. By the end of 2022, the country’s deposits amounted to 49.6 percent of GDP—in Panama, another dollarized peer, deposits stood at 117 percent of GDP.

El Salvador’s banking system was dollarized immediately, but the conversion of the circulating currency was voluntary, with citizens allowed to decide if and when to exchange their colones for dollars. Ocampo notes that, in both Ecuador and El Salvador, only 30 percent of the circulating currency had been exchanged for dollars four months after dollarization was announced so that both currencies circulated simultaneously. In the latter country, it took over two years for 90 percent of the monetary base to be dollar‐​based.

Cachanosky explains that, in an El Salvador‐​type, voluntary dollarization scenario, the circulating national currency can be dollarized as it is deposited or used to pay taxes, in which case the sums are converted to dollars once they enter a state‐​owned bank account. Hence, “there is no need for the central bank to buy the circulating currency” at a moment’s notice.

Dollarization starts with both currencies and a peg. As long as people trust that dollarization will happen at the peg, the conversion can take a while. You do not need dollars to soak up every peso on day 1. Dollarization is, above, a commitment that the peg will last for years, not a necessary commitment that the peg will last a day. 

I speculated about private borrowing at lower rates than the sovereign, once default rather than inflation is the only way out for the sovereign. This happened: 

... as Manuel Hinds, a former finance minister in El Salvador, has explained, solvent Salvadorans in the private sector can borrow at rates of around 7 percent on their mortgages while international sovereign bond markets will only lend to the Salvadoran government at far higher rates. As Hinds writes, under dollarization, “the government cannot transfer its financial costs to the private sector by printing domestic money and devaluing it.”

A nice bottom line: Ask people in Ecuador, El Salvador, and Panama what they think:

This is yet another lesson of dollarization’s actual experience in Latin American countries. It is also a reason why the vast majority of the population in the dollarized nations has no desire for a return to a national currency. The monetary experiences of daily life have taught them that dollarization’s palpable benefits far outweigh its theoretical drawbacks. 

Even more important update:

From Nicolás Cachonosky How to Dollarize Argentina  The central problem is non-money liabilities of the central bank. A detailed plan. Many other blog posts at the link. See his comment below.  

Tyler Cowen on dollarization in Bloomberg. Great quote: 

The question is not how to adopt a new currency, it is how to adopt a new currency and retain a reasonable value for the old one. 

Dollarization is easy. Hyperinflate the Peso to zero a la Zimbabwe.  Repeat quote. 

Emilio Ocampo on dollarization as a commitment device

One of the main reasons to dollarize is to eliminate high, persistent, and volatile inflation. However, to be effective, dollarization must generate sufficient credibility, which in turn depends critically on whether its expected probability of reversal is low.... 

The evidence suggests that, in the long-run, the strongest insurance against reversal is the support of the electorate, but in the short-run, institutional design [dollarization] can play a critical role.

Fifty years ago, in testimony to U.S. Congress, Milton Friedman argued that “the whole reason why it is an advantage for a developing country to tie to a major country is that, historically speaking, the internal policies of developing countries have been very bad. U.S. policy has been bad, but their policies have been far worse. ... (1973, p.127).”

In this respect, not much has changed in Argentina since. 

Craig Richardson explains how dollarization failed in Zimbabwe, a wonderful cautionary tale. Deficits did not stop, the government issued "bonds" and forced banks to buy them, bank accounts became de linked from currency. Gresham's law prevailed, the government "bonds" circulating at half face value drove out cash dollars. With persistent government and trade deficits there was a "dollar shortage." 


  1. It was time for some sense. Dollarization is no panacea, but for some countries it is worth importing at least some better institutions from abroad. Nothing has harm more Latin America than the eternal cycle of financial repression - inflation - stagnation and failed reforms.

  2. This comment has been removed by the author.

  3. Great post. Currency is currently around 2% of GDP but the BCRA also issues remunerated liabilities (~10% of GDP). I am a big fan of dollarization for Argentina but think that potentially a common South American currency would be even better:

  4. If outsourcing monetary policy is such a good idea, why outsource it to the US? There has to be better alternatives out there?

    What about NOK. Cheapest currency out there. Budget surplus and huge savings :D

    1. Do you remember last time you had a Norwegian banknote in your hands? As far as I know, they get dollars for their oil so they use dollars each time they buy something abroad...
      China has tried to "internationalise" the RMB despite a trade surplus - as expected, it does not really work.
      The US dollar is the only currency widely available around as a useful foreign currency, thanks to the US trade deficit.

    2. Whether the currency is mostly digital or physical does not matter? And almost all countries use foreign currency when importing stuff from abroad. So you think China have tried hard to internationalize RMB? Difficult to see without free flow of capital.

      I'm not seriously suggesting NOK as a global currency. I'm just saying that dollarization (USD) might not be the optimal answer to Argentina's problems.

  5. Thank you John, great piece. I agree that dollarization should be much better compared to the lousy performance of our CB (even taking into account the monetary and fiscal outlook in the US!). As you well say, sequence is the key.

    I am glad that Milei postponed two days ago the decision to dollarize as a first step in the sequence. Why it is a good decision to postpone:

    - Today you need to borrow USD to retire pesos. Bad timing, low probability, requires financial engineering while a fire is burning your house
    - The majority of prices at ARS 1000 per USD indicate that (if pesos are retired at an FX=1000) there will be an inflation in USD in the next two or three years, of at least 100%, as relative prices align. Ideally you want a quick drop in the inflation rate once you change institutions. (A good bottle of wine sells for USD7, I guess the "correct"price is USD 15/18. A haircut costs today is USD4, during our currency board was USD15). So, it is necessary to put some basic order in fiscal accounts and relative prices (including raising subsidized energy and transportation utilities) before retiring pesos at a very weak FX
    - A bimonetary system like that used in Uruguay and Peru is perhaps a cheaper and more expedite alternative to discipline the state and the CB, and closer to the experience we had with the currency board in the 1990s. This would NOT be a currency board, just currencies circulating together in various markets.

    In Uruguay, you have whole sectors operating in dollars (pricing and payments), like agriculture, cattle and forestry. Pesos and USD live together in a healthy competition, without restrictions in the FX market and legal tender for all currencies freely chosen by parties to a contract. That should be done now (with an aggressive reform of the state and deregulation), even at the risk of accelerating the fall of the peso.
    Alfredo Irigoin

  6. Your question : "Why not?"

    Your answer:

    Fed funds rate = Current Inflation Rate + Natural Rate of Interest + alpha1 * (Current Inflation Rate - Desired Inflation Rate) + alpha2 * (Potential GDP - Current GDP)

    1. Your question: "Why not?"

      Your answer:

  7. Is one option to finance the exchange for the US to lend Argentina dollars secured against future seniorage revenues? I don't know if the expanded seniorage for the US would be enough to cover it. But presumably if you secure enough years, you can get to any arbitrary amount. If so it would have a certain win-win simplicity to it.

    1. In 1998 a bill was introduced to encourage official dollarization in emerging countries by sharing seignorage with them. It makes perfect sense to do so and could provide a basis for "help" in countries like Vzla that badly need it. Additionally, stabilizing economies helps American companies who can sell more to willing customers. For more see:

  8. Re "democracy" or "democracies"

    Any alleged expert or layperson who talks about "democracies" AS IF a real democracy ACTUALLY EXISTS ANYWHERE IN THE WORLD (or has existed at any time in 'human civilization') is evidently a fool who's repeating mindlessly and blindly the propaganda fed to them since they were a kid and/or is a member of the corrupt establishment minions whose job is to disseminate this total lie because any "democracy" of 'human civilization' has always been a covert structure of the rule of a few over the many operating behind the pretense name and facade of a "democracy": (or

    "There is no America. There is no democracy. There is only IBM and ITT and AT&T and DuPont, Dow, Union Carbide, and Exxon. Those are the nations of the world today. [...]. We no longer live in a world of nations and ideologies [...]. The world is a college of corporations, inexorably determined by the immutable laws of business. The world is a business [...]." --- from the 1976 movie “Network”

    "We can either have democracy in this country or we can have great wealth concentrated in the hands of a few, but we can't have both." --- Louis Brandeis, Supreme Court Justice

    Does anyone still not see how the deadly game on the foolish public is played ... or still does not WANT to see it?

  9. The role of a central bank is to engineer optimal inflation for that country. The US Fed is doing pretty darn good for us right now (after a little slip up in 2021 that allowed a bit more over target inflation for longer that was necessary to allow relative prices to adjust to the COVID-Putin shocks. It would be a weird coincidence if the inflation that is optimal for the US and its shocks to be optimal for Argentina. [Of course, for a small economy, there is little that inflation can do to affect relative prices for good or ill, so less role for a central bank, but Argentina is a much larger economy than Panama, El Salvador or Ecuador.] Dollarization can foreclose shocks coming from fiscal policy (if it is politically stable) and its not hard to imagine eliminating the shocks of bad fiscal policy might be well worth importing US inflation which would be sup-optimal for adjusting to other Argentina idiocentric shocks.

  10. In Argentina if you want a new car or a piece of real estate, you pay in dollars. As soon as you get paid you take the pesos and get dollars. Otherwise in a few months your pesos are useless.

    Dollarization or pegging to a basket of Dollar/Euro/ Yen/Yuan is the only way out for Javier Milei and his countrymen. He cares about them as much as the Peronists and other Fascists/Communists on the other hand ruin them.

  11. Good post, thanks! I'm not so sure about this one, though:

    "Small businesses, new businesses, businesses that don't have easily valuable assets, businesses where it is too easy for the managers to hide cash, are forced to borrow, to issue debt"

    Usually, small firms with hard-to-value assets (like tech startups) tend to be financed with equity, because it's very hard for debtors to seize the assets in case of a default (trade-off theory). Restaurants are an example of something that's fairly easy to value (there's a ton of restaurants to compare to) and loans are usually collateralized with buildings or kitchen equipment.

    I also think the comparison of sovereign debt with equity is a bit of a stretch. In equity, you also have some upside risk when the company you're investing in is doing really well, which is not really the case for sovereign debt: Best case scenario you simply get all your money back with little additional upside.

  12. For those interested in reading more deeply about dollarization, here is a recently issued bibliography listing around 500 scholarly writings on the subject:

  13. I've often wondered why people regard inflation as more palatable than outright default considering the former effectively bankruptcy the country anyway. And I realized, lurking behind all of that, is the fear that outright default will lead to socialism. In the worst case, you get 30 years of Stalin. In the benign case, you get Hugo Chavez.

    1. Presumes a government must borrow.

  14. The post rightly points to some of the difficulties of dollarizing Argentina.

    Dollarization can take place in many different forms (an important topic many critics oversee). How you dollarize defines (a) how long it takes from the government's point of view and (b) how many US dollars you need.

    For the specific case of Argentina, here is non-technical summary of the proposal endorsed by Milei during his presidential campaign. This is based on my book with Emilio Ocampo.

    1. Thanks much for the comment! I added an update to the blog post. Good luck!

  15. I now know more about dollarization then i thought possible and have great hope for the success of Argentinas new President. If only our own government could get its self together and think not of themselves but my kids and grand kids. Greed is good, as long as it is not coupled with arrogance and ignorance which are in vast supply in all levels of federal state and local governments.

  16. "Zimbabwe recently dollarized, simply giving up on its currency after the greatest hyperinflation ever seen. The argument for Argentina is similar. Zimbabwe just did it."

    Mind you, Zimbabwe subsequently pulled a 180, introduced a new proto-dollar, and started down the path of de-dollarization.

  17. A quick question:
    "Interest-paying debt can simply be redenominated in dollars once you pick a rate" - these dollars should still come from somewhere, right? I have a deposit in a bank for one million peso, which should be redenominated for 1000 USD - where does these 1000 USD come from? There might also be an issue of (at least temporarily) opening the bank positions on USD, what about financial stability?

  18. Argentina’s political history is more characterized by liberalism than socialism. For example, there were six coups d'état that overthrew democratic governments during the 20th century: in 1930, 1943, 1955, 1962, 1966 and 1976. Most of these coups were supported by the oligarchy, the Catholic Church and anti-communist sectors of the armed forces. They imposed liberal policies that resulted in economic and social failures.

    In particular, recent history has been marked by extreme neoliberal policies that have led to disastrous outcomes: the last military dictatorship (1976-1983) committed a genocide and waged a hopeless war with England to try to boost its popularity; the Menem administration (1989-1999) (by the way, you are mistaken if you think that being a Peronist implies having socialist policies) that pretty much sold the country; De la Rúa government (1999-2001), which faced a massive economic and social crisis that resulted in violent repression and his resignation, literally escaping in the presidential helicopter; and the Macri presidency (2015-2019), which implemented austerity policies that worsened the country’s debt and inflation problems.

    Milei does not fit the definition of a libertarian. He is a right-wing and fascist media personality with no political background. His choice of a former neo-nazi youth member as his adviser has sparked outrage ( He has also received the support of many convicted genocidaires because, literally, he and his vicepresident advocates for the release of military personnel convicted for crimes against humanity. He has proposed to shut down CONICET, the leading science organization in Latin America, claiming that scientists are lazy and undeserving of public funding . Come on, he said the a presidential debate, in live television in front of 40 million people, that "climate change is a lie of socialism to write fake papers". He has already sued 3 journalists who accused him of being a Nazi: he not very interested in freedom of expression. He is a fascist clown.

    Right now Milei is backing down from the plan of full dollarization. He also has contradicted others of his campaign promises by appointing several former Macri officials to his cabinet. He has chosen Luis Caputo, who was the president of the central bank under Macri, as his economic minister. This is surprising, considering that Milei had previously criticized and insulted Caputo harshly, accusing him of "smoking US$ 15 billion of the reserves irresponsibly" and saying that "Caputo is an incompetent who knows nothing about economics and who got us into debt for 100 years". Milei, who ran as an independent and won with the promise of punishing previous politicians, has now given key positions to Macri’s allies. In other words, he is now a Macri puppet.

    Argentina faces a bleak future as Milei and Macri will once again sell out the country’s interests and resources (a recurring pattern every four or eight years). Argentina will sink into unprecedented levels of crisis and misery. Then a Peronist gobernment will stabilize the situation and the cycle will repeat.


Comments are welcome. Keep it short, polite, and on topic.

Thanks to a few abusers I am now moderating comments. I welcome thoughtful disagreement. I will block comments with insulting or abusive language. I'm also blocking totally inane comments. Try to make some sense. I am much more likely to allow critical comments if you have the honesty and courage to use your real name.