Research Articles and Working Papers
"Solving Open Economy Models with Incomplete markets by Perturbation"
This paper proposes to solve open economy models with incomplete markets by first approximating the local dynamics of a tractable auxiliary model, and then applying regular perturbation to some of the parameters to reach the model of interest. The method is easy to implement with available solution packages, and can approximate models around a large subset of the state-space, including the stochastic steady-state. The lead application extends the two-period, multi-asset model of Coeurdacier and Gourinchas (2016, JME) to an infinite horizon setup. The calibrated model with bonds and equities delivers a large level of equity home bias, and a natural link between trade and financial openness with external asset positions comparable to the data. On the downside, the model generates excessive risk-sharing, and counterfactual co-movements of gross capital flows.
"Consistent Approximations around the Stochastic Steady-State"
This paper combines Taylor projection with perturbation on future random shocks to generate polynomial approximations to the policy functions of dynamic stochastic models around their implied fixed-point. The approximations are fast to compute, and accurate in the vicinity of the fixed-point, which converges to the stochastic steady-state of the true model as one increases the order of the approximation. When the deterministic steady-state exists, a perturbation strategy around the standard local solution delivers the risk-corrected coefficients without the need of non-linear solvers. The method, however, is capable of solving a wide class of incomplete-markets models for which a deterministic steady-state fails to exist, including models with portfolio-choice, models with occasionally binding constraints, dynamic Heterogeneous-agent models, and discrete-choice models of strategic default.
"Approximate Aggregation Theory for Heterogeneous-Agent Models"
This paper proposes a new perturbation strategy to build a sequence of polynomial approximations to the equilibrium laws of motion of dynamic heterogeneous-agent models, with one or many assets. The point of approximation is the stationary equilibrium, and the endogenous aggregate states include the first Ith moments of the wealth distribution. As I increases, the sequence converges to the true local solution, but the marginal contribution of the higher moments decays asymptotically. In practice, approximations with a small number of moments already provide an excellent degree of accuracy, and are easier and faster to compute than the stationary equilibrium itself, in a way that parallels the solution of standard DSGE models. Moreover, the method can deliver solutions that break certainty-equivalence, allowing for risk-premia and non-trivial portfolio choice even when the deterministic stationary equilibrium does not exist.
"Solving Heterogeneous-Agent Models around the Ergodic Steady-State"
This paper develops a novel local method to solve recursive stochastic macroeconomic models as an approximation around the ergodic mean of the distribution of the variables. In contrast to standard local approaches, agents fully take risk into account up to a first-order, which helps to solve several well-known limitations of previous algorithms (need of a well-defined deterministic steady state, certainty-equivalence up to a first-order, inability to deal with incomplete markets). The method is very fast, easy to implement, and it provides an excellent degree of accuracy as measured by the Euler equation errors. We introduce the method by solving two well-known examples of incomplete-markets models in the literature: the stochastic growth model with heterogeneous agents and aggregate risk, and the New Keynesian model with heterogeneous agents (HANK).
"A Welfare State-based Fiscal Multiplier"
Empirical evidence indicates that fiscal multipliers were significantly larger during the austerity period of 2010-13 than in previous times. A common explanation is that new-Keynesian models can deliver a larger-than-usual spending multiplier when the zero lower bound binds. Here we argue that these theories cannot explain the evidence if the combination of policy-makers committing to a sizable reduction of public debt and low output growth expectations lead agents to believe that austerity measures were permanent. Instead, we explain the evidence in the context of a new-Keynesian model with incomplete markets and heterogeneous households where a permanent cut of welfare state spending at the zero lower bound increases incentives to save for precautionary reasons, thus leading to a Paradox-of-Thrift type of recession. We also provide new empirical evidence consistent with the model, as we find that the welfare state spending multiplier was significantly larger than the non-welfare and tax multipliers for advanced countries engaging in fiscal adjustments over this time period.
"The Great Recession: Divide between Integrated and Less Integrated Countries" (with Eric van Wincoop and Gang Zhang) . IMF Economic Review 64.1 (2016): 134-176.
No robust relationship has been found between the decline in growth of countries during the Great Recession and their level of trade or financial integration. Here we confirm the absence of such a monotonic relationship, but document instead a strong discontinuous relationship. Countries whose level of economic integration (trade and finance) was above a certain cutoff saw a much larger drop in growth than less integrated countries, a finding that is robust to a wide variety of controls. We argue that standard models based on transmission of exogenous shocks across countries cannot explain these facts. Instead we explain the evidence in the context of a multi-country model with business cycle panics that are endogenously coordinated across countries.