UCLA

This dissertation consists of three chapters on macroeconomics and finance. In Chapter 1, I study how disruptions in secondary bond market liquidity affect the macroeconomy. I introduce search-based secondary markets for long-term corporate bonds into a dynamic general equilibrium model. In the model, with borrowing constraints and incomplete insurance, firms restrict hiring ex-ante when default risk increases. A worsening of bond market liquidity, by affecting bond prices and thus the borrowing limits for firms, has aggregate negative impact on firms' labor choices. A positive default-liquidity spiral further amplifies these effects. In the quantitative analysis of my model, I show that a liquidity shock calibrated to match the observed increase in the bid-ask spread could explain about 20% of the employment losses in the Great Recession. I also provide a structural estimate of the impacts of the Fed's corporate bond purchasing program on the real economy during the COVID-19 crisis. By improving bond market liquidity, the Fed's interventions avoided a 2 percentage point drop in employment.

In Chapter 2 (joint with Adrien d'Avenas and Andrea Eisfeldt), we explain why credit spreads explain firm-level investment better than equity volatility does. While credit spreads always predict lower investment, the sensitivity of investment to equity volatility changes sign in the cross section of firms depending on their distance to default. Higher equity volatility predicts greater investment for firms far from their default threshold, consistent with a larger option value of investment at higher levels of volatility. On the other hand, higher equity volatility predicts lower investment for firms with high credit spreads, consistent with debt overhang. Opposite effects at the firm level wash out and confound aggregate inference. We provide clean intuition using a simple model.

In Chapter 3 (joint with Lucyna Gornicka, Federico Grinberg, and Marcello Miccoli), we study how the introduction of Central Bank Digital Currency (CBDC) might disintermediate the banking sector. Using a simple portfolio choice model we find that CBDC reduces bank credit only in special cases and when it does, the effect is quantitatively small. In the model, households allocate their wealth between an illiquid asset and three liquid assets: cash, bank deposits and CBDC. An imperfectly competitive banking sector provides deposits and lending. When all liquid assets are costless to access, the introduction of CBDC does not lead to bank disintermediation, as banks increase the return on deposits to fight off the competition from CBDC. However, if the access to deposits and CBDC is costly, the introduction of the latter may lead to bank disintermediation under specific conditions. The conditions are that CBDC is much cheaper to access than bank deposits and that the wealth distribution is very unequal. Under these conditions, poorer households will stop holding deposits in favor of CBDC, but banks will not aggressively fight the outflow of customers due to their relatively small wealth. Still, the impact on lending turns out quantitatively small if banks have access to other forms of funding.