2014 Fiscal and Monetary History of Latin America Conference
April 11–12, 2014
Featured Media Playlist
Friday, April 11
Beginning in the 1950s and 1960s, Peru, Chile, and Argentina experienced remarkably similar trajectories of political and economic upheaval. Peru and Chile are the success stories, stabilizing and enjoying strong growth since the early 1990s; Argentina was a bit of an exception, according to Saki Bigio of Columbia University,.
Bigio presented a paper investigating whether Peru and Chile were “just lucky,” or if there were important reasons why Argentina’s moves were less effective. “It’s very useful to study history and narrative of how reforms were carried out and the political influences behind them,” Bigio said. “Our contribution is to deploy economic theory to parse out why these reforms worked, or didn’t.”
“In the 1950s through the 1970s, all three nations had some form of classic Latin American socialism,” marked by nationalization of key industries, agrarian reform, and massive social programs, Bigio said. This led to massive growth in government expenditures not supported by the tax base, especially in Chile.
All three nations eventually experienced serious deficits, some form of hyperinflation, some form of default, and severe GDP decline, he showed. And all experienced violent political transitions.
Bigio, who grew up in Peru in the late 1980s and 1990s, vividly recalled the outcomes of “really bad policy”: regular power blackouts and frequent bombings, no water, stacks of pesos required for purchases, and long waits to get telephone service or a buy a car. Policy oscillated. “Every government elected seemed to reverse the reforms of the previous one, or return to policies of the past,” he recalled.
Eventually, these nations adopted very similar reforms: new constitutions, fiscal discipline, privatization, social security reform, and central bank independence. All had some form of price controls, minimum wages and tariffs that were reversed through labor and trade reform and a shift to market-based price systems.
Chile stabilized first, starting in 1979 with reforms Pinochet adopted. Fujimori made similar reforms in the 1990s and Peru has enjoyed strong and steady growth ever since.
But Argentina was less stable and suffered a severe decline in 1981 when it left currency convertibility.
Bigio postulated a very simple explanation: Argentina was never capable of implementing successful reforms—a common theme in Tom Sargent’s recent work. The difference boiled down to politics.
“For starters, we have to look at who players are. In Chile, Pinochet was a dictator with virtually no opposition. In Peru, Fujimori took a strong line to tackle terrorism of the Shining Path Maoist insurrection, which made him a popular figure. When Congress opposed his pension reforms, he simply shut down Congress.”
In Argentina, President Carlos Menem had to negotiate reforms with Congress and with strong provincial governors. Local and regional governments in Argentina had more power in determining federal representation. Provincials contributed just 13 percent of national revenue but accounted for 37 percent of federal expenditures, mostly associated with transfers and pensions.
In all three nations, the social safety net of entitlements emerged with the ability to tax, and was used for political leverage, creating a fiscal burden. All three privatized, but in different forms.
“The way to think of this is a pecking-order theory of populism funding,” Bigio concluded. “When each nation ran out of resources, they nationalized resources. When they ran out of those resources, they nationalized pension systems. When they ran out of those resources, they nationalized everything via inflation. “
Over time, there was a similar cascade of reversing reforms.
Discussant Manual Amador of Federal Reserve Bank of Minneapolis said the paper raised the question of what constitutes a successful and sustainable reform. He said it was useful to distinguish between those that only change the state variable and reforms that also change the driving forces in the economy.
He presented a model of an open economy and the political environment, with budget and sustainability constraints that showed that running more debt leads to more economic distortion. Temporary reforms result in temporary economic improvements; it takes political strength to implement the slow, often painful process of permanent reform.
Amador noted the importance of learning, where agents could see benefits of reform and learn from past experiences and neighbors, and the possibility of external incentives to implement reform.
Conference participant Nancy Stokey pointed out a lack of distinction in the work between between policies and political structures. “Learning about policies that work elsewhere doesn’t do much good without political structure reform,” she noted.
Despite being one of the poorest countries being examined as part of fiscal history project, Bolivia still faced a historical timeline similar to most Latin American nations: revolution, crisis and hyperinflation, heavy nationalization of industry, stabilization and finally growth.
Carlos Gustavo Machicado, presenting his work with Tim Kehoe, said that the pivot points of each of these periods could be traced to shocks both internal and external to the Bolivian economy, with government policy playing an increasingly dramatic role in the direction of the economy as the nation grew older.
The role of debt in Bolivian government evolves along a similar timeline. “During nearly all of its economic history, the Bolivian government ran chronic deficits," said Machicado. But in the mid-1980s, a hyperinflation crisis made it very difficult for Bolivian authorities to borrow from private creditors, turning instead to bilateral and eventually multilateral creditors to avoid crisis. Since most of this foreign credit came in the form of dollars, the Bolivian economy rapidly moved toward using the dollar, reaching up to 90 percent of all day-to-day transactions. More significantly, as the Bolivian government depended more heavily on foreign lending, they became increasingly bound to honoring foreign debt obligations over time, for fear of losing access to their shrinking sources of external credit.
Government officials moved through several cycles of reform to attempt to reduce their debt through repayment and renegotiation. By 2006, the Bolivian government had nearly all their debt wiped clean, and they were able to stabilize their economy by taking a leading role in the economy via state-owned companies working in the production of domestic oil, electricity, and telecommunications.
Discussant Fernando Alvarez saw some suspect growth accounting in Bolivia’s data. He noted that large amounts of aggregation of all state-owned enterprises made it difficult to discern successes from failures in public investment. Attendees debated the issue at length, but the consensus was expressed by co-author Tim Kehoe. “What’s lacking, and people noticed, is better information on public enterprise deficits. But I don’t think we’re going to get better data.” However, he and Machicado remain hopeful that a new, large dataset direct from the Bolivian government may contain the detail they need to better understand growth in the later periods of Bolivian fiscal history.
A decade of hyperinflation propped up by wage and price indexing stands out as the central feature of Brazil's economic history, said Márcio Garcia, presenting work from himself, Diogo Guillén, and Patrick Kehoe. Indexation was a "second best option" to price stability, in the words of Milton Friedman, and the policy allowed Brazil to weather hyperinflation without the expected output collapse; however, the policy also may have allowed that protracted period of inflation to persist.
Five attempts to stabilize inflation failed to take root politically, since the real economy was insulated from the effects of hyperinflation and the people of Brazii had yet to feel the effects; at last, the appropriately-named "Real Plan" took hold in 1994, stabilizing inflation up to today.
Garcia argued that those five previous plans–each adjusting, freezing or automating changes to wages, prices, exchange rates and interest rates in varying degrees between 1980 and 1994–largely failed to stabilize the Brazilian economy because none of them stopped the core problem: the unfettered printing of new currency. While this "Real Plan" did not have the most fruitful changes to fiscal policy compared to previous periods, the seigniorage rate–where a central banking authority makes money by issuing new currency–went from 3.5 percent of the GDP down to 0.7 percent. That change had a dramatic impact on inflation, supporting a narrative of a tight relationship between the growth of the money supply and decline of inflation.
That, in the view of the authors, makes the Real Plan as implemented in Brazil an example consistent with Friedman's reasoning that monetary policy stands as a far more effective anti-inflationary tool than fiscal policy.
In discussing the their conclusions, Rodolfo Manuelli didn't think the accounting looked quite so simple. "The question is really whether the adjustment needs to happen contemporaneously, or whether it can happen four years in the future," said Manuelli, noting that the specific details of how the government structured its budgets were less important than how their actions played out in the documented rate of inflation.
Specifically, looking at the data, Manuelli said that the demand for money did not appear as stable as the authors believed; he noted that this fact undermines elements of the accounting exercise that the authors set out in the paper. He also questioned whether indexation alone could explain why hyperinflation lasted as long as it did in Brazil. Manuelli ultimately compared Brazil to a runner preparing who runs a marathon without a rigorous training schedule leading up to the race; to him, the Real Plan tells a story of a rapid decrease in inflation without the accompanied lead up, at least by the accounting as it stands in the authors' work.
Other attendees noted that this discrepancy could potentially be explained by a closer look at the central bank's balance sheet. That said, Thomas Sargent marveled that the authors had clearly uncovered an anomaly in the known history of Brazil's economy, even if missing data makes the full picture murky. “These Hidden Markov Models are a bit of magic," said Sargent. "There’s stuff you don’t see, but you see their shadows.”
Like nearly all the Latin American nations discussed during the conference, Chile's battle with hyperinflation went on for extended periods, divided into distinct periods marked by differing fiscal and monetary policy curatives. In presenting his and Diego Saravia's work examining Chilean fiscal history, Rodrigo Caputo noted that, as in the case of many Latin American countries, these periods of economic policy upheaval is tied inextricably with the political upheaval Chile experienced between 1960 and today. In that time frame, Chile had seven nonconsecutive democratically-elected governments, with an intervening 17-year authoritarian rule under Augusto Pinochet; the coauthors set out to establish exactly how the fiscal, monetary and debt management policies under each regime affected macroeconomic outcomes.
Inflation rose to more than 700 percent shortly after Pinochet seized power, despite his attempts to slash the country's fiscal deficit. Caputo pointed to the removal of price controls in September 1973 as a potential explanation. At any rate, Pinochet's economists were able to craft policy that slowly drove the inflation rate back down over the course of his rule, leaning heavily on seigniorage to finance government expenditures while maintaining a fiscal deficit of nearly zero.
In 1979, a fixed exchange rate was adopted, which in turn induced trade balance deficits and reduced the cost of foreign borrowing. Public debt decreased rapidly as private debt increased. Inflation dropped to single digit levels by 1981.
However, by 1982, the fixed exchange rate was abandoned, driving up foreign interest rates and making it impossible for Chilean banks to service their own debt. Central bank rescue programs and liquidations followed the crisis, bringing the portfolios of private banks into the hands of the Chilean Central Bank. In the ensuing two decades, even in the transition back to a democratic government, Chile avoided default by maintaining a policy of absorbing the cost of crises in its own treasury and central bank, increasing government debt in favor of stability. The newly nationalized debt was indexed long term, in foreign currency to avoid being monetized by lendors.
Discussant Andrew Powell re-emphasized the authors' implication about the link between economics and politics. “The region has quite a history, and this project is a great opportunity to celebrate and investigate that." But examining those correlations can also complicate economic analysis, Powell noted, pointing to the assertion that for Chile, "default was not an option." That seems more a moral stance than economically sound reasoning, and Powell said that it stands out in a discussion of economic tools. He also suggested some more detail on the rates of return of government instruments would help clarify the effectiveness of individual instruments, rather than the collective policy action in a given period.
Alejandro Hernandez-Delgado, Felipe Meza, and Ignacio Trigueros-Legarreta–the authors examining Mexico's fiscal history–were unable to attend the conference, so discussant Gerardo della Paolera presented their basic thesis before launching into an extended discussion of their work.
In the paper, the authors use the work of Sargent, Kareken and Wallace and conference organizer Juan Pablo Nicolini as a guide to trace the economics of budget constraints between 1978 and 2009. Like many other Latin American countries, Mexico held far more of its debt in foreign currency compared to nominal domestic debt, particularly in the mid-1980s. But launching into the discussion early, Timothy Kehoe pointed out that in this case, the distinction between foreign and domestic debt isn't all that meaningful, since the data only records where the debt was sold, not who holds it over time.
Rising fiscal deficits drove that uptick in sovereign debt over the 1970s and 1980s, reaching a crescendo in 1982 when the authors postulate that the political pressure to halt escalating indebtedness pushed the economy into an inflationary scenario.
By 1990, the Bank of Mexico had been given independence to pursue all possible avenues for maintaining the purchasing power of the peso, including the issuing of Tesobonos (bonds issued in pesos but indexed to the U.S. dollar) and substituting short term borrowing with a sudden stop. The shift to tight monetary policy seemed unsustainable due to the exploding interest on short term debt already held by Mexico, particularly when the change in macroeconomic policy directive had to weather political uncertainty in the wake of the assassination of Luis Donaldo Colosio in 1994. Political turbulence in Mexico during this timeframe had a significant impact on interest rates and Mexico's international reserves.
Paolera noted that it appears unclear whether debt accrued in this period was expected to be repaid, since a huge amount of the lending was done by development banks; he suggested that the authors looked into it further.
The discussion concluded with attendees examining Latin America more broadly in the 19th century, before any of its countries had central banks of any kind. Even then, fiscal and monetary policy inconsistency was harming growth in the region. It's possible that the policy trope of printing money to alleviate budget deficits took root in the policymaking tradition of Mexico and many other nations far before they were so deeply influenced by foreign economies.
Drawing parallels from fiscal crises in Mexico, Timothy Kehoe presents model for analyzing the sovereign debt crises of 2010-2012 in the Eurozone. The need to sell large quantities of bonds every period leaves the government vulnerable to self-fulfilling crises in which investors, anticipating a crisis, are unwilling to buy the bonds, thereby provoking the crisis. The optimal policy is to reduce debt to a level where crises are not possible, unless the nation is in a recession. In that case, with potential of future increase in revenues, the government may optimally choose to "gamble for redemption," running deficits and increasing its debt, but also increasing its vulnerability to crises.
Saturday, April 12
Can a sovereign debt crisis be result of self-fulfilling equlibria? Juan Pablo Nicolini explored that question in a presentation of joint work with Gaston Navarro and Pedro Teles of Banco de Portugal.
Nicolini noted a feedback loop where countries facing high interest rates run a higher probability of default, which in turn causes lenders to charge even higher rates, and default becomes even more likely.
For example, leading up to Argentina’s 2001 default, the nation faced an average interest rate of 10 percent, against an average debt-to-GDP ratio of 30 percent. Argentina’s interest payment was equivalent to the paying the 3 percent rate on U.S. bonds against 100% GDP ratio. “The question in our minds at the time was, if we would have had 3 percent instead of 10 percent rate, would we have defaulted?”
Ultimately, the International Monetary Fund withdrew support for Argentina and the nation defaulted three months later. By comparison, in 2012, Spain, Italy, and Portugal ran GDP ratios of 80 to 100 percent and yet did not default, because the European Central Bank intervened.
In the paper presented, Nicolini and coauthors studied the effects of the policy of deep-pocket lenders such as the IMF or ECB. They take into account that lenders will offer a varying schedule of interest rates depending on the debtor nation’s overall debt level.
Their work builds on a Calvo (1988) model in which the interest rate schedule is contingent on the market value rather than the face value of debt. They show that Arellano (2008) exhibits the multiplicity, or multiple solutions, seen in the Calvo model.
They characterize the equilibria in a two-period model. They describe and compare the differences between the Calvo and Arellano rate schedules. “The schedules are the same with just a change of variables, but it makes a big difference to an agent whether he is facing one or another,” Nicolini noted.
Finally, he showed that using an infinite-period model, a sunspot can be used to replicate behavior of spreads in Italy and Spain during a sovereign debt crisis.
During the presentation and discussion by Chicago Booth’s Veronica Guerrieri, conference participants raised numerous questions and issues about the model. Nicolini acknowledged the skepticism, noting that when manipulating the model with the sunspot, “we can get pretty much anything we want; that’s the source of the skepticism.”
Peru’s modern fiscal and monetary history falls into three broad periods that Marco Vega outlined in his talk.
The 1960s and ’70s were marked with a large fiscal expansion but no tax increases, leading to unprecedented levels of public debt, followed by hyperinflation when that debt hit a ceiling. The economy was based on export of primary products, mostly minerals, but dwindling supply of natural resources and lack of investment limited growth.
Pent-up demand for education, public services, and redistribution to address inequality fueled the economic policies of the “dependency school,” popular among politicians, intellectuals, and the military academy. This included expanded public spending, attempts at land reform, and government takeover of industries.
The 1980s, a period Vega called “heterodox Peru,” were characterized by experiments. After an initial drop in inflation, the government stopped servicing its debts in 1985. This was combined with price controls, loose credit, and continued public spending. These policies, inspired by the idea that inflation is not a monetary phenomenon, spiraled into hyperinflation and the worst crisis in Peruvian history.
The 1990s were marked by efforts at price stabilization: demise of price controls and generalized subsidies, Central Bank independence, and the introduction of market-oriented reforms.
Vega shared a series of graphs showing the outcomes of these policies over the decades. Inflation ran as high as 400 percent monthly and was not controlled until the 1990s. Exchange rates, fixed until 1975, underwent some massive devaluations in the ’70s and ‘80s.
Vega pointed out that in contrast with other stabilization episodes in region, Peru’s monetary anchor was not the exchange rate but the base money growth rate.
Inflation fell very slowly compared countries using exchange rate stabilization.
Since the late 1990s, Peru has enjoyed stability and growth. The challenge for economists is to understand why. “What has changed in Peru since 1996? Why are the same or very similar politicians behaving more responsibility? Is this social learning?” Vega wondered.
In lieu of a discussion of Vega’s presentation on Peru, Diego Restuccia presented a similar short overview of Venezuela’s economy.
Oil has been both a blessing and a curse for Venezuela, his data showed. When oil was discovered there in 1941, multinationals arrived to drill and brought know-how and development, building roads and creating towns.
But the oil that built the nation’s infrastructure was later mismanaged. The oil industry was nationalized in 1978, but the oil earnings were all taxed away. Paradoxically, there were periods during oil booms where revenues increased, yet government borrowing also rose.
Public debt began rising steadily in 1974 and was at its worst in 1985-95, with huge spikes in inflation. Attempts at reforms in the late ’80s and early ’90s were somewhat successful, but didn’t last long after an initial coup attempt by Hugo Chavez in 1992. When Chavez took power in 1999, inflation picked up again.
Overall, Venezuela has experienced high volatility and declining economic performance. In 1960, its per-capita gross domestic product relative to the US was .9; it has declined ever since, all the way to .25 by 2000.
Wrapping up the conference, Restuccia noted that most Latin American economies are not doing that well in terms of productivity and GDP growth. In 1960, the region’s relative GDP per capita was .30 compared to the US. By 2009, it had declined to .23.