My dissertation investigates what the main sources of fluctuations in some are of
key macroeconomic variables and how they propagate in the economy. More importantly, I quantify these shocks by using statistical methods and study their transmission mechanism in both closed and open economy settings.
The first essay, “Explaining International Business Cycles with Real Frictions and
Technology Shocks: A Bayesian Approach”, uses a Bayesian approach to estimate a
standard international real business cycle model augmented with time non-separable
preferences, variable capacity utilization and investment adjustment costs. First, I
find while most of the output and consumption fluctuations attribute to neutral and
world technology shocks, investment-specific technology (IST) shocks explain the bulk
of investment volatility. Second, my estimated model with IST shocks simultaneously
accounts for the negative correlation between the real exchange rate and relative consumption and the negative correlation between the terms of trade and relative output.
IST shocks act as demand shocks by giving firms an incentive to increase labor and
capacity utilization and households an incentive to increase domestic absorption, thus
causing appreciation of international prices when consumption and output are also
rising. In addition, compared to other real shocks, IST shocks generate high volatilities
in the relative prices. Third, by using marginal likelihood comparison, I find that the success of the model depends to a large extent on variable capacity utilization;
investment adjustment costs play only a minor role. After a positive IST shock,
increased capital utilization both induces a stronger response of labor by increasing
the marginal product of labor and generates comovement of consumption and output.
The second essay, “Monetary Policy and the Wealth Effect”, studies the transmission
of monetary policy shocks under different wealth effects on labor hours. In
our simple sticky price model with no capital, I find that the responses of both real
and nominal variables, such as consumption, labor, marginal cost and inflation, are
highly volatile and have very little inertia under preferences which imply a zero wealth
effect on labor hours. Under a zero wealth effect, wage responses are very high, which
further causes high volatility in marginal costs and inflation. This result is robust
to inclusion of any degree of backward indexation in inflation, wage rigidity, capital
or endogenous capital utilization rate. Furthermore, using a Bayesian approach
I estimate a small-scale dynamic stochastic general equilibrium model augmented
with time non-separable preferences, a sufficiently large number of real and nominal
frictions and structural shocks. First, I find that the estimated time non-separable
preferences imply near-zero wealth effect on labor supply. Second, in our variance
decomposition exercise I find that policy shocks explain the bulk of volatilities in
variables, confirming the calibration results. Finally, by using marginal likelihood
comparison, I find that the data rejects the model with weak wealth effect preferences
in favor of preferences with habit formation which can generate reasonable
inertial responses of inflation to a policy shock.