My research focuses on topics in empirical macroeconomics and monetary economics as well as applied microeconomics. In particular, I use relatively novel approaches to study how the inflation targeting (IT) framework and other monetary policy issues affect macroeconomic outcomes. Much of my applied microeconomic research considers how policies and other shocks affect labor markets, financial behavior, and health outcomes. Click on the titles below to view my papers.
Job Market Papers
This paper asks whether inflation targeting (IT) "just got lucky" in the sense that its adoption in a number of countries in the 1990s just happened to coincide with a more tranquil macroeconomic environment or whether it can be credited with contributing to a more stable macroeconomic structure. There is wide disagreement about the macroeconomic effects of inflation targeting, both theoretically and empirically. I contribute to the debate by using an innovative approach that allows the generation of counterfactual variances of inflation and output that would have prevailed if IT were not introduced. In particular, I use the counterfactual Vector Autoregressive (VAR) method developed by Stock and Watson (2002) to determine whether the observed decline in the variability of inflation and output growth can be attributed to a more stable structure (the propagation mechanism) or less violent shocks (the impulses). I find that the observed moderation in inflation volatility may be attributed largely to a more stable structure associated with the introduction of IT and less to the consequence of milder shocks. I estimate the propagation mechanism to account for more than 50 percent of the decline in inflation volatility for the majority of countries in the sample. Meanwhile, the observed tranquility in the business cycle is driven solely by much less violent shocks which have offset what appears to be a less stable structure seemingly arising from the IT framework. Results provide indirect evidence that the achievement of low and stable inflation may have come at the expense of output stabilization, as predicted by theory.
While there is a general consensus that reforms are needed to improve the financial viability of the Social Security program, these reforms may have welfare implications --- both intended and unintended --- that remain unexamined. I study the 1983 Social Security reforms to examine how a large, unanticipated negative retirement wealth shock affects workers over their lifecycle. I exploit the nonlinearity in the reform design to estimate causal effects on the labor supply and on the savings of workers at different stages of the lifecycle. My findings suggest that men responded to the policy change by altering their labor supply, though only when they were near the normal retirement age. Women increased their labor supply, both at the extensive and intensive margins. In addition, I show that workers responded through higher savings prior to retirement. The Social Security amendments appear to have disproportionate effects, with some lower-educated workers remaining in the labor force in later years. Enhancing public understanding of the implications of future reforms could mitigate potential adverse effects particularly on vulnerable subpopulations.
Works In Progress
Is the quest for 'miracle drug' over?: Initial Evaluation of the Impact of Truvada on HIV Rates
Since its approval for pre-exposure prophylaxis (PrEP) intended for uninfected people, Truvada has been touted as a 'miracle drug' that could potentially slow down the spread of HIV on a global scale. This paper is the first to examine the effect of the increased utilization of Truvada on HIV diagnosis rates. Given the efficacy of Truvada in preventing users from acquiring the virus, one might expect HIV diagnosis rates to decline as PrEP intake rises over time. However, Truvada does not provide 100 percent protection. Ex-post moral hazard in the form of risky sexual behavior, poor adherence to medication, and the threat of drug-resistant strains are all potential reasons why the increased use of Truvada may not necessarily result in lower HIV rates. I employ instrumental variables-fixed effects model on state-level data to identify the effect of Truvada use on HIV rates. Results show that the increased intake of Truvada appears to have raised HIV diagnosis rates, after controlling for state-invariant factors and aggregate time effects. In particular, an increase in the number of Truvada prescriptions by 1,000 raises mean HIV rate by about 1.2 percentage points. Meanwhile, a million dollar increase in Medicaid reimbursement of Truvada increases the average HIV rate by about 1.03 percentage points. This provides suggestive evidence of the existence of moral hazard and poor adherence for some PrEP users as well as drug-resistant HIV strains. The quest for miracle drug for HIV may not be over after all.
Firm-specific Pay Premia and Academic Honors: Evidence from administrative data (with Nhu Nguyen)
Do students who obtain academic honors sort to firms with higher pay premium? Using administrative data on firms and workers matched to administrative college enrollment data, we provide estimates on the impact of getting academic honors on firm-specific pay premia. First, we decompose individual wages into individual-specific component (fixed and time-varying) and to firm-specific wage premium. Then, we take advantage of the GPA cutoffs in determining academic honors to address unobservable characteristics associated with obtaining the honors. In particular, we use regression discontinuity design that compares graduates who just made it past the cutoff to get academic honors and those who almost did. This research provides evidence on the extent to which workers with academic honors are matched with higher-paying firms. It also provides insights on whether firms respond to signals associated with academic honors.
Retirement, Health Investments, and Cognitive Decline among older Americans
This research revisits how retirement affects health investments and age-related cognitive decline of older Americans. Empirically, it has been shown that retirement could lead to either better or worse health outcomes. I contribute to the existing literature by estimating the causal effect of retirement on health investments (or “disinvestments”) in terms of physical activity and preventative health tests on the one hand, and smoking and drinking on the other. I also contribute to the literature on the effect of retirement on cognitive functioning where there is a huge variation in both the sign and magnitude of the empirical estimates. Using a panel dataset which tracks Americans age 50 onwards, I employ difference-in-difference IV models to disentangle the effect of retirement on health. As retirement is endogenous to health and likely to health investments as well, I exploit the 1983 Social Security reforms (which raised the retirement age and reduced prospective lifetime wealth of workers born in 1938 or later) as a source of exogenous variation that affects workers’ retirement decisions. First, I estimate the effect of the Social Security reforms on the probability of retirement on the affected cohorts. Then, I use the estimated retirement probability as an instrument for actual retirement status in the second stage where the outcomes are the various health investments and cognition measures.
Impact of Data Revisions on Monetary Policy Formulation: Evidence from OECD countries
Economic data series are revised typically when more information becomes available, estimation methodologies are improved, or as compilation standards are upgraded over time. In the monetary policy-making process, monetary authorities examine the latest available information on a wide array of variables to assess domestic and external developments and form a forward-looking assessment of inflation pressures. It could happen that policy actions implemented on the basis of real-time information could differ considerably from what would have been the recommendation if the final data had been used. Hence, revisions in economic data can potentially complicate the policy-making process. This article focuses on the potential impact on monetary policy formulation in economic data among OECD countries. In particular, it examines whether the interest rate prescribed by a monetary policy rule would differ if the final data are used instead of the preliminary data.
"Surveillance of the Philippine Rice Market," BSP Economic Newsletter 14:2, (2014).
"Assessing Inflation Vulnerability: An Early Warning System for Inflation in the Philippines" (with Dennis Mapa), Theoretical and Practical Research in Economics 4(8), (2013):137-152.
"Tenets of Effective Monetary Policy in the Philippines" (with Jasmin Dacio), BS Review 14:1, (2012):14-30.