Asset Pricing, Monetary Policies, and the Zero Lower Bound

The Great Recession challenged the conventional way of conducting monetary policy and sparked debates on optimal monetary policy in the presence of the zero lower bound on nominal interest rates, and its interaction with asset prices in financial market. My dissertation shows how the zero lower bound changes the conduct of monetary policy, and how considering asset prices in monetary policy helps to improve
the welfare of the economy.

In Chapter 1, I first examine the role of money in precautionary saving and the behaviour of other asset prices, such as returns on bonds and stocks. Money is introduced via the form of transaction cost into a production economy with limited stock market participation, where agents with a lower inter-temporal elasticity of substitution, called "non-stockholders", have no access to the stock market. This
model not only quantitatively resolves the risk premium puzzle and the risk-free return puzzle and matches volatilities of key macroeconomic variables, but also corresponds with empirically documented facts regarding money growth, inflation, and asset prices in the literature. I then examine whether money growth should respond to equity prices and equity premiums. I find that monetary policy improves welfare for both stockholders and non-stockholders if it reduces equity premiums in the economy. The model thus prescribes money growth rules that are pro-cyclical with respect to equity prices or equity premium changes.

Chapter 2 studies how the conduct of monetary policy affects the frequency of the zero lower bound binding in a canonical New-Keynesian Dynamic Stochastic General Equilibrium (DSGE) model. The model generates a much deeper recession if the zero lower bound binds than when it does not. The model economy almost never hit the zero lower bound if monetary policy is described by a Taylor rule with interest rate smoothing. Increasing the central bank's reaction to deviations of inflation and output does not increase the probability of the economy hitting the zero lower bound. With interest rate smoothing, a lower inflation target significantly increases the likelihood of the economy hitting the zero lower bound. I also show that linear approximation solutions fail not only quantitatively but also qualitatively in analyzing the zero lower bound.

The final chapter explores the role of credibility and what optimal policy the central bank can credibly make for forward guidance by solving for the whole set of sustainable sequential equilibria (SSE) in a standard New Keynesian model with an occasionally binding constraint on the nominal interest rate. Under full commitment, forward guidance is a longer duration of nominal rates at zero even when the contractionary shock disappears, followed by a quick revert-to-normal path. However, under the best SSE, there are many different policy paths that can be interpreted as forward guidance. One possible policy path features a smaller decrease of the nominal rate when the contractionary shock happens and an even longer duration at the zero lower bound than under full commitment. Another policy path keeps the nominal rate away from zero all the time and results in smaller inflation. The economy reverts more slowly to its normal state under the best SSE. The key insight is that central banks with same credibility may choose very different policies in their forward guidance and they do not have to lower nominal rate to zero. The full commitment equilibrium is not generally implementable.