Theses Doctoral

Essays on Micro-Level Consumption Behavior and Open Economy Macroeconomics

Hong, Seungki

This dissertation finds significant differences in the micro-level household consumption behavior between emerging and developed economies, disentangles multiple possible explanations for these differences, and evaluates their macroeconomic implication on business cycles.

The first chapter estimates the marginal propensity to consume (MPC) out of transitory income shocks using micro data for an emerging economy. To this end, I employ a nationally representative Peruvian household survey. Two striking differences emerge when the Peruvian MPC estimates are compared with U.S. MPC estimates obtained by the same method. First, the mean MPC of Peruvian income deciles is much higher than that of U.S. deciles. Second, within-country MPC heterogeneity over the deciles is substantially stronger in Peru than in the U.S.

The second chapter studies the driving factor for the MPC differences between Peru and the U.S. I begin by exploring three possible explanations for the stronger MPC heterogeneity in Peru through the lens of a standard precautionary saving model: liquidity constraints, consumption front-loading behavior, and heterogeneous interest rates. Then, I disentangle these possible explanations by examining relevant data patterns appearing in the micro data. Specifically, participation rates in borrowing activities and consumption growth rate patterns of the income deciles suggest that liquidity constraints drive the stronger MPC heterogeneity in Peru. Then, I decompose the cross-country MPC gap into the component driven by liquidity constraints and the component caused by factors unrelated to liquidity constraints. To this end, I delineate a top income group unaffected by liquidity constraints in each country by conducting an MPC homogeneity test and evaluate its MPC. I find that liquidity constraints are also important for explaining the higher mean MPC in Peru.

The third chapter makes a first attempt to study emerging market business cycles in a heterogeneous-agent open economy model. A central question in open economy macroeconomics is how to explain excess consumption volatility in emerging economies. This chapter argues that to understand this phenomenon, it is important to take into account households' idiosyncratic income risk, precautionary saving, and MPCs. Financial frictions determining asset liquidity in the model are calibrated such that MPCs are as high as empirical estimates from Peruvian micro data, which are substantially greater than the U.S. MPC estimates. I then estimate the model using macro data and Bayesian methods. The model captures the observed excess consumption volatility well. To highlight the importance of high-MPC households in driving this result, I show that excess consumption volatility disappears when households are counterfactually replaced with those exhibiting U.S. MPCs. High-MPC households contribute to consumption volatility through i) their strong consumption response to resource fluctuations and ii) large consumption reduction when assets become more illiquid. The transmission mechanisms of trend shocks and interest rate variations that previous studies use to explain excess consumption volatility are dampened because households significantly deviate from the permanent income hypothesis, on which these mechanisms crucially depend.

Geographic Areas


  • thumnail for Hong_columbia_0054D_16485.pdf Hong_columbia_0054D_16485.pdf application/pdf 2.07 MB Download File

More About This Work

Academic Units
Thesis Advisors
Uribe, Martin
Ph.D., Columbia University
Published Here
April 20, 2021