Working Papers
1. Long-Term
Securities and Banking Crises, with Zhouxiang Shen
and Dongling
Su, PDF,
October 2023
Abstract: The US bank holdings of long-term securities
have increased in recent years. As is witnessed by the recent bank failures
including SVB, prices of long-term securities are sensitive to interest rate
hikes and can trigger bank runs. To study the role of bank holdings of
long-term bonds, we incorporate banks in a DSGE framework. We study how
cost-push shocks and the associated passive or active
interest rate hikes affect the macroeconomy including inflation, investment,
and output. The procyclical bank balance sheets and long-term bond prices
amplify the adverse shocks, which can trigger bank runs. We introduce two types
of macroprudential policies that can mitigate or prevent banking crises: a
permanent tax on bank holdings of long-term bonds and a cyclical tax that
responds to interest-rate changes.
2. Fiscal
Stimulus under Average Inflation Targeting, with Zheng Liu and Dongling Su, PDF, November 2022
Abstract: The stimulus effects of expansionary fiscal
policy under average inflation targeting (AIT) depends on both monetary and
fiscal policy regimes. AIT features an inflation makeup under the monetary
regime, but not under the fiscal regime. In normal times, AIT amplifies the
short-run fiscal multipliers under both regimes while mitigating the cumulative
multiplies due to intertemporal substitution. In a zero-lower-bound (ZLB)
period, AIT reduces fiscal multipliers under a monetary regime by shortening the
duration of the ZLB through expected inflation makeup. Under the fiscal regime,
AIT has a nonlinear effect on fiscal multipliers because of the absence of
inflation makeup and the presence of a nominal wealth effect.
3. Robust Inattentive Discrete Choice, with Lars Hansen and Hao
Xing, PDF, February 2022
Abstract: We introduce robustness to the rational inattention model
with Shannon mutual information costs in a discrete choice setting when the
decision maker is concerned about model misspecification/ambiguity. We provide
necessary and sufficient conditions for the robust solution and develop
numerical methods to solve it. We show that the decision maker slants their
beliefs pessimistically toward worse outcomes. As a result, their choice
behavior can be qualitatively different from that in the standard rational
inattention model with risk aversion. We apply our model to three consumer
problems and show that tests based on misspecified
models can lead to type I errors.
4. Robust
Financial Contracting and Investment, PDF,
with Aifan Ling and Neng Wang, Jaunary 2021.
Abstract: We study how investors' preferences for robustness influence corporate
investment, financing, and compensation decisions and valuation in a financial
contracting model with agency. We characterize the robust contract and show
that early liquidation can be optimal when investors are sufficiently ambiguity
averse. We implement the robust contract by debt, equity, cash, and a financial
derivative asset. The derivative is used to hedge against the investors'
concern that the entrepreneur may be overly optimistic. Our calibrated model
generates sizable equity premium and credit spread,
and implies that ambiguity aversion lowers Tobin's q, the average investment,
and investment volatility. The entrepreneur values the project at an internal
rate of return of 3.5% per annum higher than investors do.
5. Multivariate
LQG Control under Rational Inattention in Continuous Time. February 2019
Abstract: I propose a multivariate linear-quadratic-Gaussian control framework
with rational inattention in continuous time. I propose a three-step solution
procedure. The critical step is to transform the problem into a rate distortion
problem and derive a semidefinite programming representation. I provide
generalized reverse water-filling solutions for some special cases and
characterize the optimal signal dimension. I apply my approach to study a
consumption/saving problem and illustrate two pitfalls in the literature.
Abstract: Stocks with
high exposure to oil price movements perform well when oil price rises and
poorly when oil price falls. This oil driven stock price momentum differs from
the conventional stock price momentum and it is
related to firms’ profitability. Daily oil price movement measures the state of
the economy at high frequency and exposures to oil price movement estimated
with daily data can be significant in forecasting future returns.
Abstract: We study a continuous-time principal-agent problem with
learning under ambiguity. The agent takes hidden actions to affect project
output. The project quality is unknown to both the principal and the agent. The
agent faces ambiguity about mean output, but the principal does not. We show
that incentives are delayed due to ambiguity. While belief manipulation due to
learning about unknown quality causes wages and pay-performance sensitivity to
be front-loaded, ambiguity smooths wages and causes the drift and volatility of
wages to decline more slowly over time. When the level of ambiguity is
sufficiently large, the principal fully insures the
agent by allowing the agent to shirk forever.
Abstract: We incorporate long-term defaultable corporate bonds and credit risk in a dynamic stochastic general equilibrium business cycle model. Credit risk amplifies aggregate technology shocks. The debt-capital ratio provides a new state variable and its endogenous movements provide a propagation mechanism. The model can match the persistence and volatility of output growth as well as the mean equity premium and the mean risk-free rate as in the data. The model implied credit spreads are countercyclical and forecast future economic activities because they affect firm investment through Tobin's Q. They also forecast future stock returns through changes in the market price of risk. Finally, we show that shocks to the credit markets are transmitted to the real economy through Tobin's Q.
The following papers are outdated and need revisions. But they are still cited by a few papers.
1.
Stationary
Equilibria of Economies with a Continuum of Heterogeneous Consumers, PDF, 2002
2.
Managerial Preferences, Corporate
Governance, and Financial Structure, PDF, with Hong
Liu, March 2006
3.
Experimentation under
Uninsurable Idiosyncratic Risk: An Application to Entrepreneurial Survival, PDF, with Neng Wang, May 2007
Abstract: We propose an analytically
tractable continuous-time model of experimentation in which a risk-averse
entrepreneur cannot fully diversify the idiosyncratic risk from his business
investment. He makes consumption/savings and business exit decisions jointly,
while learning about the unknown quality of the project over time. Using the
closed-form solutions, we show that (i) the
entrepreneur may stay in business even though the project's net present value
(NPV) is negative; (ii) entrepreneurial risk aversion erodes option value and
lowers private project value so that a sufficiently risk-averse entrepreneur
may exit even when the NPV is positive; (iii) a more risk-averse or a more
pessimistic entrepreneur exits earlier; and (iv) the model can generate a
positive relation between wealth and entrepreneurial survival duration from
undiversifiable idiosyncratic risk without liquidity constraints.
4.
Corporate
Tax Policy and Long-Run Capital Formation: The Role of Irreversibility and
Fixed Costs, PDF, May 2008.
Abstract: This paper presents an
analytically tractable continuous-time general equilibrium model with
investment irreversibility and fixed adjustment costs. In the model, there is a
continuum of firms that are subject to idiosyncratic shocks to capital.
Although the presence of investment frictions lowers consumer welfare, it may
raise or reduce the long-run average capital stock, depending on the degree of
idiosyncratic uncertainty. An increase in this uncertainty may raise
equilibrium aggregate capital, but reduce welfare. An
unexpected permanent change in the corporate income tax rate affects the
investment trigger and target values, and hence the size and rate of capital
adjustment. Following this tax policy, the percentage changes in equilibrium
quantities are larger when fixed adjustment costs are larger. These changes are
significantly smaller in a general equilibrium model than in a partial
equilibrium model.