A BDSGE Model of Singapore for Policy Experiments and Analysis

Abstract

This paper estimates a small open economy New Keynesian DSGE model forthe Republic of Singapore. The model is adapted from the work of Galí and Monacelli(2005) and Galí (2015), and draws from the main literature precedent of aBDSGE model for Singapore from Chow, Lim, and McNelis (2014). It introducesreal frictions in the form of habit persistence, nominal frictions in the form of pricestickiness, an exogenous preference shifter, and an exchange-rate targeting policyrule. The model is estimated using Bayesian inference and produces posteriorpoint estimates for the model’s parameters, as well as impulse response functionsthat describe the effect of exogenous structural shocks on observable macroeconomicvariables. Overall, parameter estimates are in line with the literature onboth the Republic of Singapore and small open economies in Asia, and the impulseresponses behave in line with expectations. In addition, we conduct modelcomparisons of the baseline model against several other variations, including analternative specification of a CPI inflation targeting Taylor rule monetary policy,a domestic inflation targeting rule, and a model without habit formation. Modelcomparison results find a strong preference in the data for the inclusion of habitformation behavior, and also for a CPI inflation based Taylor rule policy specification.We exercise caution when interpreting these results by acknowledging thepresence of identification problems in estimating BDSGE models, and by conductingan identification analysis of our model parameters. Although the possibilityremains that some parameters are only weakly identified, we find that our modelreasonably estimates the macroeconomic behaviors of Singapore’s economy.