Dollarization in Argentina

Dollarization means “the total elimination of the Argentine currency, the peso, and its complete replacement with the U.S. dollar,” with transactions, accounts, debts, and contracts all redenominated and the dollar becoming “the sole legal tender.” The authors frame it as the most extreme case of two nested debates: internationally, the choice of exchange-rate anchor and optimal currency areas, and domestically – since international considerations “matter less” for a relatively closed economy like Argentina’s – the choice between monetary rules and discretion, of which “dollarization is the ultimate rule, or the total absence of discretion.” They note that Argentina is “already partly dollarized,” with 61.3 percent of private nonfinancial-sector deposits denominated in dollars, so complete dollarization “would not dramatically change” existing habits.

Argentina’s currency has been unstable “stretching back to the nineteenth century,” with paper money not made convertible into gold until 1899, convertibility suspended in 1914 and again in 1929, and a central bank established in 1935 that began “monetizing deficits” from 1943 on – after which the price level rose by a factor of 10 in the U.S. but by 10^12 in Argentina. High-inflation episodes recurred in 1975, 1982-85, and 1987-90, culminating in an inflation rate of 78 percent per month when Carlos Menem took office in May 1989. To end this, Congress passed the March 1991 convertibility law pegging the austral (replaced by the peso in January 1992) to the dollar at a fixed rate, alongside fiscal reforms including reduced spending and taxes, privatization, trade liberalization, and financial deregulation.

Under a currency board, “a country’s currency is fully backed by foreign reserves”; Argentina’s central bank is forbidden by its 1992 charter from lending to the government and must maintain free reserves – gold, foreign currency, or deposits and bonds payable in gold or foreign currency – of at least 100 percent of the monetary base. The main departure from the classic form is that the central bank may also hold Argentine government bonds as backing, but the authors characterize this as “minor” because those bonds must be purchased at market price (not as direct loans to the government), cannot exceed 33 percent of total reserves, and cannot grow by more than 10 percent in any year – and as of November 1999 only 3 percent of reserves were actually held this way.

The price level, which had been exploding, “was quickly stabilized and has remained stable” after 1991, and annual GDP growth rose from a 1 percent average in 1980-90 to 4.3 percent in 1991-98, reversing more than the 23 percent fall in real per capita output suffered during the 1980s. This expansion was interrupted twice: by the 1995 “Tequila effect” following Mexico’s January 1995 devaluation, during which total bank deposits fell 13 percent even though the composition between peso and dollar deposits barely changed, and by the 1998-99 “Vodka-Caipirinha effect” following Russia’s August 1998 default and Brazil’s January 1999 devaluation, despite Argentina’s limited direct trade exposure to either country (Mexico took only 1.7 percent of Argentine exports; Brazil-bound exports were 3 percent of Argentine GDP).

Unilaterally, Argentina need only “eliminate the peso-denominated monetary base”: since commercial banks already hold dollar-denominated reserves rather than peso deposits at the central bank, the monetary base is simply currency in circulation, which Argentina has more than enough liquid reserves ($19.0 billion against a $16.5 billion base, as of end-1999) to buy back and exchange for dollar notes. A bilateral alternative would have the Federal Reserve print roughly $14 billion in new notes to hand over to Argentina in exchange for retiring the peso, letting Argentina keep its own reserves (and the interest income on them) in escrow as collateral against ever reintroducing a national currency – an arrangement the authors judge to face real practical and political difficulties, not least because “it is not clear what advantages the U.S. could draw from such an association,” which would also require U.S. congressional approval.

Seigniorage is the income a central bank earns because “its liabilities (money) bear no interest, while its assets do”; only about $700 million of Argentina’s $17.2 billion average 1998 liquid reserves represented true seigniorage on the $14.9 billion monetary base, equal to roughly 0.2 percent of GDP, or 1.2 percent of government revenues. Capitalized as a permanent annual flow using a formula relating the current monetary base to the nominal interest rate, the real growth rate, and the real interest rate, this comes to roughly six times the current monetary base (about $84 billion) under assumptions of 6 percent nominal interest, 4 percent real interest, and 3 percent growth – though the authors caution this present-value figure “is very sensitive to the assumptions about rates of interest and growth rates.”

Fixed exchange rates, including currency boards, “always present credibility problems and are subject to self-fulfilled speculative attacks,” because if enough investors believe a devaluation is coming, their rush to convert pesos to dollars can itself deplete the reserves and force the devaluation they feared. Despite the currency board’s legal backing, the authors note the Argentine executive retains “emergency powers that would allow it to suspend convertibility immediately by decree, subject to ratification by Congress after the fact,” and point to Argentina’s history of six coups (1930, 1943, 1946, 1951, 1966, 1976) as reasons investors’ residual devaluation fears, evident in the Tequila and Vodka-Caipirinha episodes, are not irrational. Dollarization “would provide a much stronger commitment device,” especially bilaterally, since reintroducing a national currency would be far harder to do quietly if an international treaty explicitly prohibited it.

The authors grant that dollarization “takes away that ability from central banks” to create liquidity on demand, but point to mechanisms Argentina built after the 1995 Tequila crisis – itself “in large part a run on the Argentine banking sector” – that substitute for discretionary money creation: dollar-denominated bank reserve requirements (21 percent of deposits by October 1999, versus 1.3 percent in the U.S.), a deposit insurance fund, and a contingent repurchase facility with 14 foreign banks that can expand the monetary base by up to 50 percent on demand. Combined, these mechanisms “provide Argentina with protection for about 39 percent of its deposits… or more than 2.4 times the monetary base,” which the authors compare favorably with the U.S. Federal Reserve’s 1.3 percent weekly monetary-base expansion after the 1987 stock market crash, concluding the facility “alone” gives Argentina lender-of-last-resort capacity comparable to or exceeding a conventional central bank’s.

The authors note that Argentina’s past exercise of independent monetary policy produced only “disastrous consequences,” so the relevant comparison is not between dollarization and “a good independent policy” but between dollarization and Argentina’s actual policy record; moreover, Argentina “has already made the decision” to give up discretionary policy by adopting the currency board, so it “is thus not a question of choosing between a rule and discretion, but rather of reaping the benefits of a choice that it has already made.” Box 1 formalizes the underlying “time-commitment problem” (Kydland and Prescott 1977): with a government minimizing a loss function over inflation and unemployment given a Phillips curve driven by unanticipated inflation, a government that moves after private expectations are set is always tempted toward positive inflation, yielding the same equilibrium unemployment “natural rate” as full commitment but with higher inflation – and “dollarization is a way to reduce” the government’s feasible inflation choices “to the single point {0}.”

Treating the 0.2 percent of GDP annual seigniorage cost as an insurance premium against Tequila-type shocks (which the authors estimate caused a permanent 14 percent output loss), the breakeven annual probability of such a shock is only 1.4 percent; since Argentina “has been hit twice in ten years” by comparable shocks, the authors conclude “unilateral dollarization is unambiguously desirable under those assumptions.” They are nonetheless explicit that this is a rough calculation resting on a specific, simplified model of crisis risk, and the paper’s conclusion draws back from a firm recommendation: “we do not reach a definite answer on whether Argentina should dollarize,” and Argentina “must weigh very carefully the consequences of losing the ability of pursuing an independent monetary policy” before doing so, since a bad past record does not guarantee that better independent policy could never be achieved in the future.