I am an Analyst at the Office of the Superintendent of Financial Institution, Canada's banking and insurance regulator. I hold a Ph.D. in Economics from the University of Toronto.
I was previously a Research Assistant at the Bank of Canada, and a Financial Engineer at d1g1t Inc, a FinTech startup offering a cutting-edge reporting and analytics platform to wealth managers.
In my academic work, I developed econometric tools to analyze the industrial organization of banking with respect to systemic risk and financial stability. My working papers explore various aspects of regulatory impacts on banking and lending.
In my latest research paper, I research how the participation of non-banks in the lending market may increase the systemic risk in the economy. Using a structural model of the syndicated loans market, I show that blanket capital requirements on all lenders, though it encourages banks to decrease their risk exposure, may encourage non-bank lenders to take greater risks. This effect is driven by the increase in the cost of capital due to the regulation, and the different risk-aversion preferences of non-banks when compared to banks. I presented a poster version of this paper at the 2021 American Finance Association conference. I would have presented at the CEA and at the SCSE in May 2020, but the world went on lockdown.
In another paper, titled Time-Varying Elasticity of Substitution in Near-Money Assets (presented at the 29th Annual Meeting of the Midwest Econometrics Group), I show that the Liquidity Coverage Ratio (LCR) mandated by Basel III might be misguided due to some varying elasticity of substitution between money and safe assets used to cover that liquidity. Understanding how some assets are substitutes for money gives us a better sense of how sensitive the economy might be to potential liquidity shocks.
I also published a paper in collaboration with Martin Burda. In this paper, we develop and analyze the performance of a Constrained Hamiltonian Monte Carlo method to estimate a Copula Multivariate GARCH process. We find benefits for efficient inference on complicated constrained dependence structures when compared to a traditional random-walk Markov Chain Monte Carlo methodology.