"A Decade Lost and Found: Mexico and Chile in the 1980s"with Raphael Bergoeing, Patrick J. Kehoe, and Raimundo Soto.
Chile and Mexico experienced severe economic crises in the early 1980s. This paper analyzes four possible explanations for why Chile recovered much faster than did Mexico. Comparing data from the two countries allows us to rule out a monetarist explanation, an explanation based on falls in real wages and real exchange rates, and a debt overhang explanation. Using growth accounting and a calibrated growth model, we find that the difference in the performance of Chile and Mexico was driven not by differences in inputs of capital and labor stressed by traditional theories of depressions, but rather by differences with which these inputs were used, measured as total factor productivity (TFP). Using economic theory to interpret historical evidence, we conclude that the crucial difference between the two countries was earlier reforms in government policy in Chile. The most crucial of these reforms were in banking and bankruptcy procedures. (Description of the data and MS Excel file with all of the data.)
"Great Depressions of the Twentieth Century"with Edward C. Prescott.
(This paper is the introduction to the volume that includes the first paper; it lays out the research methodology and summarizes the results of the whole research project.)
The papers in this volume study twelve great depressions—those in Canada, France, Germany, the United Kingdom, and the United States in the interwar period; those in Argentina, Brazil, Mexico, and Chile during the "lost decade" of the 1980s; those in New Zealand and Switzerland that began in the early 1970s; and another great depression in Argentina that began in 1998. It also studies three not-quite-great depressions—that in Italy in the interwar period and those in Finland and Japan in the 1990s. All of the papers rely on growth accounting to decompose changes in output into the portions due to changes in factor inputs and the portion due to the changes in efficiency with which these factors are used. All of the papers employ simple applied dynamic general equilibrium models. Collectively, these papers indicate that government policies that affect productivity and hours per working-age person are the crucial determinants of the great depressions of the twentieth century. (Description of the data and MS Excel file with all of the data.)
"Is Switzerland in a Great Depression?"with Kim J. Ruhl.
Abrahamsen, Aeppli, Atukeren, Graff, Müller and Schips (2005) object to Kehoe and Prescott's (2002) characterization of the Swiss economy as being in a great depression over the period 1974-2000. They argue that (1) depressions should be defined in terms of declines in labor productivity rather than in GDP; (2) examining deviations from trend in GDP is equivalent to examining levels; (3) Swiss data from the 1970s should be ignored because it is of low quality and because the 1970s were a period of turmoil in the Swiss labor market; (4) Swiss GDP data should be adjusted to account for appreciations in the terms of trade; and (5) the change in Swiss national accounts from a system based on SNA68 to one based on SNA93 will make Swiss economic performance look better. In this note, we find that none of these arguments have merit except for, possibly, the need to adjust GDP data for changes in the terms of trade. We conclude that Switzerland has indeed suffered a great depression and, in fact, is mired in it even today. (Description of the data and MS Excel file with all of the data.)
"Modeling Great Depressions: The Depression in Finland in the 1990s"with Juan Carlos Conesa and Kim J. Ruhl.
This paper is a primer on the great depressions methodology developed by Cole and Ohanian (1999, 2007) and Kehoe and Prescott (2002, 2007). We use growth accounting and simple dynamic general equilibrium models to study the depression that occurred in Finland in the early 1990s. We find that the sharp drop in real GDP over the period 1990-93 was driven by a combination of a drop in total factor productivity (TFP) during 1990-92 and of increases in taxes on labor and consumption and increases in government consumption during 1989-94, which drove down hours worked in Finland. We attempt to endogenize the drop in TFP in variants of the model with an investment sector and with terms-of-trade shocks but are unsuccessful. (Description of the data and MS Excel file with all of the data.)
"Policy-Driven Productivity in Chile and Mexico in the 1980s and 1990s"with Raphael Bergoeing, Patrick J. Kehoe, and Raimundo Soto.
(This short paper summarizes the analysis in the first paper.)
Both Chile and Mexico experienced severe economic crises in the early 1980s, but Chile recovered much faster than did Mexico . Using growth accounting and a calibrated dynamic general equilibrium model, we conclude that the crucial determinant of this difference between the two countries was the faster productivity growth in Chile , rather than higher investment or employment. Our hypothesis is that this difference in productivity was driven by earlier policy reforms in Chile , the most crucial of which were in banking and bankruptcy procedures. We propose a theoretical framework in which government policy affects both the allocation of resources and the composition of firms.
"Recent Great Depressions: Aggregate Growth in New Zealand and Switzerland"with Kim J. Ruhl.
Throughout the 1950s and 60s real GDP per working-age person in New Zealand and Switzerland grew at rates at or above the 2 percent trend growth rate of the United States. Between 1973 and 2000, however, real GDP per working-age person in both countries has fallen a cumulative 30 percent below the trend growth path. Our growth accounting attributes almost all of the changes in output growth to changes in the growth of total factor productivity (TFP), and not to changes in labor or capital accumulation. A calibrated dynamic general equilibrium model that takes TFP as exogenous can explain almost the entire decline in relative output in both New Zealand and Switzerland . To understand the recent growth experiences in New Zealand and Switzerland , it is necessary to understand why TFP growth rates have fallen so much. (Description of the data and MS Excel file with all of the data.)
"Using the General Equilibrium Growth Model to Study Great Depressions: A Reply to Temin"with Edward C. Prescott.
Three of the arguments made by Temin (2008) in his review of Great Depressions of the Twentieth Century are demonstrably wrong: that the treatment of the data in the volume is cursory; that the definition of great depressions is too general and, in particular, groups slow growth experiences in Latin America in the 1980s with far more severe great depressions in Europe in the 1930s; and that the book is an advertisement for the real business cycle methodology. Without these three arguments — which are the results of obvious conceptual and arithmetical errors, including copying the wrong column of data from a source — his review says little more than that he does not think it appropriate to apply our dynamic general equilibrium methodology to the study of great depressions, and he does not like the conclusion that we draw: that a successful model of a great depression needs to be able to account for the effects of government policy on productivity. (Description of the data and MS Excel file with all of the data.)
"What Can We Learn from the 1998-2002 Depression in Argentina?"
In 1998-2002, Argentina experienced what the government described as a "great depression." Using the "Great Depressions" methodology developed by Cole and Ohanian (1999) and Kehoe and Prescott (2002), we find that the primary determinants of both the boom in Argentina in the 1990s and the subsequent depression were changes in productivity, rather than changes in factor inputs. The timing of events links the boom to the currency-board-like Convertibility Plan and the crisis to its collapse. To gain credibility, the Argentine government took measures to make abandoning the plan more costly. Because the government was unable to enforce fiscal discipline, however, these increased costs failed to make the plan more credible and instead made the crisis far worse when it failed. (Description of the data and MS Excel file with all of the data.)
2. Capital Flows and Real Exchange Rates
"Are Shocks to the Terms of Trade Shocks to Productivity?"with Kim J. Ruhl.
International trade is frequently thought of as a production technology in which the inputs are exports and the outputs are imports. Exports are transformed into imports at the rate of the price of exports relative to the price of imports: the reciprocal of the terms of trade. Cast this way, a change in the terms of trade acts as a productivity shock. Or does it? In this paper, we show that this line of reasoning cannot work in standard models. Starting with a simple model and then generalizing, we show that changes in the terms of trade have no first-order effect on productivity when output is measured as chain-weighted real GDP. The terms of trade do affect real income and consumption in a country, and we show how measures of real income change with the terms of trade at business cycle frequencies and during financial crises. (Description of the data and MS Excel file with all of the data.)
"Capital Flows and Real Exchange Rate Fluctuations Following Spain's Entry into the European Community"with Gonzalo Fernández de Córdoba.
Spain 's 1986 entry into the European Community was followed by a dismantling of restrictions on international capital flows. Initial trade deficits and real exchange rate appreciation were followed by trade surpluses and real exchange rate depreciation. We analyze Spain 's financial liberalization using a dynamic general equilibrium model with a traded and nontraded good where a capital poor country opens itself to its capital rich neighbors. A calibrated model has trouble accounting for the large changes in relative prices observed given the small changes in quantities. Variants of the model with frictions in factor mobility between sectors fare better.
"Real Exchange Rate Movements and the Relative Price of Nontraded Goods"with Caroline M. Betts.
We investigate the relationship between a measure of relative price of nontraded goods to traded goods across countries and the bilateral real exchange rate in a sample of 52 countries and 1326 bilateral pairs over the period 1980-2000. Studying both deviations in levels and yearly changes, we find that directional movements of the two series are highly positively correlated but that fluctuations in the relative price of nontraded goods are smaller than those of the real exchange rate. Variance decompositions say that about one-third of deviations in levels of the real exchange rate, and about one-fifth of yearly changes, can be accounted for by fluctuations in the relative price of nontraded goods. The relation between the two series is much stronger, the more important is the trade relation between the countries, and the smaller is the variance of their bilateral real exchange rate. In these results, there is no bias resulting from the presence of rich country-poor country bilateral pairs, nor from the presence of high inflation country-low inflation country pairs. (Description of the data and MS Excel file with all of the data and MS Excel file with bilateral trade data.)
"Sudden Stops, Sectoral Reallocations, and the Real Exchange Rate"with Kim J. Ruhl.
A sudden stop of capital flows into a developing country tends to be followed by a rapid switch from trade deficits to surpluses, a depreciation of the real exchange rate, and decreases in output and total factor productivity. Substantial reallocation takes place from the nontraded sector to the traded sector. We construct a multisector growth model, calibrate it to the Mexican economy, and use it to analyze Mexico's 1994-95 crisis. When subjected to a sudden stop, the model accounts for the trade balance reversal and the real exchange rate depreciation, but it cannot account for the decreases in GDP and TFP. Extending the model to include labor frictions and variable capital utilization, we still find that it cannot quantitatively account for the dynamics of output and productivity without losing the ability to account for the movements of other variables. (Description of the data and MS Excel file with all of the data.)
"Tradability of Goods and Real Exchange Rate Fluctuations"with Caroline M. Betts.
We develop a simple, multicountry, multisector intertemporal general equilibrium model in which the degree of tradability of output differs across sectors. Tradability is determined both by the degree of substitutability in consumption between units of the same good produced in different countries and by the transactions costs that must be incurred to consume goods outside their country of origin. Home bias is endogenously determined. A vector of country specific shocks is realized at each data, and there are complete contingent claims markets. A calibrated version of the model replicates well the observed relationship between movements in the bilateral real exchange rate between Mexico and the United States and movements in the relative price of comparatively nontraded goods to traded goods across countries. In addition, the shocks induce movements in trade balances and real exchange rate that are consistent with the data. Finally, the model can also match evidence on sectoral deviations from the law of one price.
"U.S. Real Exchange Rate Fluctuations and Relative Price Fluctuations"with Caroline M. Betts.
This paper investigates the relation between the U.S. bilateral real exchange rate with five of her most important trade partners and the associated bilateral relative price of nontraded goods. Traditional theory attributes fluctuations in real exchange rates to changes in the relative prices of nontraded to traded goods. Some recent research suggests that there is little or no relation between the real exchange rate and this relative price measure. We find that these negative results are not robust and, in particular, that this relation depends crucially on the choice of trade partner and on the choice of price indices used to measure relative prices. The relation is stronger the more important is the trade relationship between the United States and a trade partner. The relation is stronger when we measure relative prices using producer prices rather than consumer prices. (Description of the data, MS Excel file with all of the original data, and MS Excel file with all of the final data.)
3. General Equilibrium Theory
"Bankruptcy and Collateral in Debt Constrained Markets"with David K. Levine.
Typical models of bankruptcy and collateral rely on incomplete asset markets. In fact, bankruptcy and collateral add contingencies to asset markets. In some models, these contingencies can be used by consumers to achieve the same equilibrium allocations as in models with complete markets. In particular, the equilibrium allocation in the debt constrained model of Kehoe and Levine (2001) can be implemented in a model with bankruptcy and collateral. The equilibrium allocation is constrained efficient. Bankruptcy occurs when consumers receive low income shocks. The implementation of the debt constrained allocation in a model with bankruptcy and collateral is fragile in the sense of Leijonhufvud's "corridor of stability," however: If the environment changes, the equilibrium allocation is no longer constrained efficient.
"Liquidity Constrained Markets versus Debt Constrained Markets"with David K. Levine.
We compare two different models in a common environment. The first model has liquidity constraints in that consumers save a single asset that they cannot sell short. The second model has debt constraints in that consumers cannot borrow so much that they would want to default, but is otherwise a standard complete markets model. Both models share the features that individuals are unable to completely insure against idiosyncratic shocks and that interest rates are lower than subjective discount rates. In a stochastic environment, the two models have quite different dynamic properties, with the debt constrained model exhibiting simple stochastic steady states, while the liquidity constrained model has greater persistence of shocks.
"Lotteries, Sunspots, and Incentive Constraints"with David K. Levine and Edward C. Prescott.
We study a prototypical class of exchange economies with private information and indivisibilities. We establish an equivalence between lottery equilibria and sunspot equilibria and show that the welfare and existence theorems hold. To establish these results, we introduce the concept of the stand-in consumer economy, which is a standard, convex, finite consumer, finite good, pure exchange economy. With decreasing absolute risk aversion and no indivisibilities, we prove that no lotteries are actually used in equilibrium. We provide a simple numerical example with increasing absolute risk aversion in which lotteries are necessarily used in equilibrium. We also show how the equilibrium allocation in this example can be implemented in a sunspot equilibrium.
4. Trade Theory
"An Evaluation of the Performance of Applied General Equilibrium Models of the Impact of NAFTA"
This paper uses data to evaluate the performances of three of the most prominent multisectoral static applied general equilibrium models that were used to predict the impact of the North American Free Trade Agreement. These models drastically underestimated the impact of NAFTA on North American trade. Furthermore, the models failed to capture much of the relative impacts on different sectors. Ex-post performance evaluations of applied GE models are essential if policy makers are to have confidence in the results produced by these models. Just as importantly, such evaluations help make applied GE analysis a scientific discipline in which there are well-defined puzzles and clear successes and failures for alternative theories. Analyzing sectoral trade data indicates that a new theoretical mechanism for generating trade in the models is needed, a mechanism in which large increases in trade can take place in product categories with little or no previous trade. To capture changes in macroeconomic aggregates, the models need to be able to capture changes in productivity.
"Demographics in Dynamic Heckscher-Ohlin Models: Overlapping Generations versus Infinitely Lived Consumers"with Claustre Bajona.
This paper contrasts the properties of dynamic Heckscher-Ohlin models with overlapping generations with those of models with infinitely lived consumers. In both environments, if capital is mobile across countries, factor price equalization occurs after the initial period. In general, however, the properties of equilibria differ drastically across environments: With infinitely lived consumers, we find that factor prices equalize in any steady state or cycle and that, in general, there is positive trade in any steady state or cycle. With overlapping generations, in contrast, we construct examples with steady states and cycles in which factor prices are not equalized, and we find that any equilibrium that converges to a steady state or cycle with factor price equalization has no trade after a finite number of periods.
"How Important is the New Goods Margin in International Trade?"with Kim J. Ruhl.
We propose a methodology for studying changes in bilateral trade due to countries exporting goods that they did not export previously, or exported only in small quantities. Applying this methodology to countries pairs that undergo trade liberalization and to pairs where one of the countries undergoes significant structural transformation, we find large increases on this extensive — or new goods — margin. Looking at country pairs with no major trade policy change or structural change, however, we find little or no increases on the extensive margin. Studying time series on trade by commodity, we find that data from before 1988 and 1989, when most major trading countries moved to the Harmonized System, are not compatible with data from afterwards. (Zip file with all of the data.)
"Trade, Growth, and Convergence in a Dynamic Heckscher-Ohlin Model"with Claustre Bajona.
In models in which convergence in income levels across closed countries is driven by faster accumulation of a productive factor in the poorer countries, opening these countries to trade can stop convergence and even cause divergence. We make this point using a dynamic Heckscher-Ohlin model — a combination of a static two-good, two-factor Heckscher-Ohlin trade model and a two-sector growth model — with infinitely lived consumers where international borrowing and lending are not permitted. We obtain two main results: First, countries that differ only in their initial endowments of capital per worker may converge or diverge in income levels over time, depending on the elasticity of substitution between traded goods. Divergence can occur for parameter values that would imply convergence in a world of closed economies and vice versa. Second, factor price equalization in a given period does not imply factor price equalization in future periods.