Research

My research interest are in Empirical Banking, International Macroeconomics and Financial Economics.

Papers

• Job Market Paper: TARP Effect on Bank Lending Behaviour: Evidence from the last Financial Crisis, with Stefano Puddu
  Click to read Abstract

  Using a unique data set based on US commercial banks and county level loan origination for the period 2005-2010, we measure whether banks that benefited from the Troubled Asset Relief Program (TARP) increase small business loan originations. We propose an identification strategy which exploits the ownership structure of bank holding companies. We find that TARP banks provide on average 19% higher small business loan originations than NO TARP banks. The disaggregated data allows us to control for the potential demand side effects. When considering poverty and unemployment rates at a county level we show that TARP is effective only in counties suffering from unemployment. Several robustness checks confirm the main result.

• TAF Effect on Liquidity Risk Exposure, with Stefano Puddu
  Click to read Abstract

  Using a unique bank-level dataset, we compare the liquidity and liability features of US commercial banks according to whether they received credit from the Term Auction Facility (TAF) program during the recent financial crisis. Moreover, we identify bank features affecting the likelihood of receiving TAF support and assess the impact of the TAF program on bank liquidity risk. We control for potential selection bias by using the change in the US housing price index at state levels between 2002:Q1 and 2006:Q3 as an exclusion restriction. The results suggest that, on average, banks that benefited from the TAF program exhibited higher ex ante levels of liquidity risk measures and illiquid collateral. TAF banks drastically reduces their funding liquidity risk positions in the periods after the first time they received the financial support. These banks exhibited larger reductions in liquidity exposures and the larger the amount of reserves they received, the greater the impact. Finally, we find that TAF banks are more likely to be headquartered in US states that experienced sharper housing price appreciation before the beginning of the crisis. Several robustness checks confirm the main results.

• Liquid Assets in a Cash-in-Advance Model
  Click to read Abstract

  I construct a model where both money and a fraction of real assets can be used to purchase consumption goods, and are therefore considered as liquid. I investigate how this set up of competing media of exchange affects the static allocation of real assets and I document the dynamic response of the endogenous variables to shocks. I find that asset holdings increase when a larger fraction of the asset can be used for transactions, and that the effect increases with inflation. Also, with higher inflation, the liquidity premium of the asset increases. The dynamic response of the real variables remains close to the standard model for most variables, but the nominal interest rate reacts much stronger.

Work in Progress

• Over-Investment and Quantitative Esing in a New Monetarist Model
  Click to read Abstract

  I extend the Lagos-Wright (2005) model to include an additional asset. The asset is not a direct substitute for money, but it is tradeable after consumers discover their type. I show under which circumstances there is over-investment into the asset and it is being traded at a liquidity premium. I then introduce an exogenous shock to the return of the asset and show how the central bank can improve welfare by buying assets with newly printed money, that is, by engaging in a strategy akin to quantitative easing.

• Originate-To-Distribute: A Dynamic Contracting Approach, with Cyril Monnet
  Click to read Abstract

  We consider the originate-to-distribute business model of banking through the lens of dynamic contracting. We find that the declining bank's retention share observed in the data is a feature of the optimal contract. Also, depending on the cost to provide incentives, investors may prefer to buy shares in relatively bad projects, thus providing a possible rational for the observed (relative) underperformance of loans that were sold. Our result implies that the originate-to-distribute model is not necessarily inefficient.

Other

During my PhD, there are several things that struck me. All are linked to central banking and the financial crisis and somewhat related to my work.

The first two observations regard the significant expansion of the balance sheets of central banks in Switzerland and elsewhere. I therefore decided to track the balance sheet size of the Swiss National Bank (SNB) over time (see SNB Balance Sheet) and compare the balance sheet size to the economy's size (see CB Balance Sheet and GDP).