Department of Economics
University of Maryland
College Park, MD 20742

Graduate Program:

Undergraduate Program:


Reallocation and Productivity during Commodity Price Cycles (Job Market Paper) [pdf]

This paper investigates how low-frequency commodity price fluctuations trigger a reallocation process that endogenously generates a decline in manufacturing productivity. I build a model in which firms with heterogeneous productivity decide between two technologies with different capital intensities and choose whether to become exporters. During a commodity boom, exporters lose market share due to exchange rate appreciation. Moreover, a commodity boom increases the relative cost of capital, which is used intensively in resource production, leading to additional reallocation within manufacturing from more capital intensive to less capital-intensive manufacturing firms. I calibrate the model to the Chilean economy and show that it can match the relevant micro and macro moments. When fed with a realistic commodity price cycle, the baseline model generates about half of the productivity decline observed in the data, a figure that is two times larger than in a counterfactual economy with no technology decision.

Quantifying the role of financial frictions during the Great Recession [pdf]

The Great Recession spurred a new wave of models with financial frictions aimed to understand the particular features of such a crisis. Most of these models, however, can be mapped into prototype economies with “agnostic” intertemporal wedges, which are deemed to be not promising to explain U.S. business cycles (Chari, Kehoe, and McGrattan, 2007; CKM). Was this time different? Are these new wave of models well-suited to quantitatively account for the so called financial crisis? To answer these questions, I augment a real business cycle model with financial intermediaries that face an endogenously determined balance sheet constraint (Gertler and Karadi, 2011; GK). In order to capture the intrinsic nonlinear nature of the crisis and to give the friction the best chance to play a role, I allow for the financial constraint to bind only occasionally. This way I capture the idea of infrequent financial crises nested within typical business cycles without relying on unrealistically large shocks. I show that the model with microfounded friction is equivalent to a prototype economy with an (exogenous) intertemporal investment wedge, which is a function of the key endogenous variables associated with the friction in the baseline model. Consistent with CKM, I confirm that these type of frictions are unimportant to account for U.S. macroeconomic fluctuations over the five decades previous to the crisis. More surprisingly, I show that the CKM result is robust to (a) the extension to the Great Recession period, (b) the introduction of a nonlinear framework able to switch between “tranquil times” and “financial crises”, (c) the solution and filtering of structural shocks using nonlinear techniques, and (d) the introduction of spread data to inform the model about the severity of the friction.

Presented at: Midwest Macroeconomics Meetings, 2017; SCE 23rd Computing in Economics and Finance Conference, 2017.

The PPP puzzle and the Balassa-Samuelson effect revisited: Evidence from a panel of 45 countries [pdf]

I estimate a long-run relationship between the real exchange rate (RER) and a set of underlying fundamentals using panel cointegration techniques for a sample of 45 countries. Unlike other studies, I found strong evidence for the Balassa-Samuelson effect for the full set of countries as well as for subsets of developed and developing countries. The size and significance of the Balassa-Samuelson effect is robust to the inclusion of a series of other long-run drivers of the real exchange rate, in particular, the terms of trade, the net foreign asset position, the level of government spending, and the degree of trade openness. I further estimate an error correction model for the growth rate of the RER which sheds light on the largely studied purchasing power parity (PPP) puzzle. In particular, the model predicts significant short-run dynamics of the RER which, at the same time, is consistent with a high degree of persistence of its deviations from the long-run cointegrating relationship.