alt text 

Padma Sharma
PhD candidate in Economics

Email: padmas[at]
Address: UCI Department of Economics
3151 Social Science Plaza
Irvine, CA 92697-5100

My research interests span the union of two fields - Econometrics and Financial Economics and involve the development of Econometric methods that address research questions pertaining to banking and financial stability. I will be graduating in June 2019 from the University of California Irvine with a PhD in Economics. I am currently on the job market and will be at the ASSA meetings at Atlanta between January 4-6, 2019.

In my job market paper , I assess the responses of U.S. regulators in the banking and Savings and Loans (S&L) industries to widespread failures by comparing observed responses with benchmarks from theoretical models of optimal failure resolution.

I was a dissertation intern at the Federal Reserve Bank of New York during June-September 2018 and a recipient of the Graduate Dean’s Dissertation Fellowship in UCI over Summer and Fall 2018.

I have worked as a quantitative risk modeler and team manager in financial services prior to commencing my doctoral studies at UC Irvine. My research is informed by my insights from evaluating bank portfolios and experience in developing statistical models to manage credit and operational risk over the period spanning the Great Recession.

Research Interests: Econometrics, Financial Economics, Computational Statistics and Macroeconomics


An overview of my working papers and research projects in progress.

Working Papers

Assessing Regulatory Responses to Troubled Banks (Job Market Paper)

How can the public assess the performance of regulators who administer the resolution of troubled banks? Economic theory indicates that regulators best serve the public interest when they act to discourage moral hazard and preserve channels of financial intermediation. I study the resolution of failed financial institutions in the US over the period 1984-1992, characterized by concurrent crises in the banking and Savings and Loans(S&L) industries. I determine whether regulators in the two industries conformed to norms of optimality by developing a Bayesian estimation algorithm to uncover their decision rules. The results show that the banking regulator provided assistance and carried out liquidations with higher probability when failures occurred amid adverse and benign local economic conditions respectively. Whereas the former’s actions align with theoretical recommendations, the S&L regulator did not distinguish across failures based on underlying economic characteristics. These findings provide insights into the differences in responses that contributed to the resilience of the banking regulator through the crisis and the eventual insolvency of the S&L regulator. The estimation approach developed here addresses unobserved heterogeneity, allows straightforward inference on quantities of interest and offers a coherent framework for model comparison.

Suspension of Payments and their Consequences with Christoffer Koch, Qian Chen and Gary Richardson

Governors suspended payments by healthy financial institutions in their states four times in the last forty years. Suspensions in Nebraska (1983), Ohio (1985), and Maryland (1985), which occurred during economic expansions, had little measurable impact at the state-level. The suspension in Rhode Island (1991), which occurred during a recession, lengthened and deepened the contraction relative to states with comparable pre-suspension patterns of economic activity. We document the suspension’s impact on macroeconomics aggregates, such as unemployment, payrolls, per capita incomes, business formation, bankruptcies, and mortgage defaults, using methods of synthetic control. Our estimates help us to address two policy questions. Were the costs of direct federal and state expenditures to prevent suspension of payments from the Great Depression until today worth the benefits? Was the cost of federal interventions to prevent suspensions of payments by commercial banks following the collapse of Lehman brothers in September 2008 worth the likely benefits? In both cases, our answer is yes.

Bayesian Latent Class Modeling with Ordinal Response Data

Bayesian methods to estimate latent class models have been developed and implemented in contexts involving continuous and binary outcome data but have not yet been adapted to model ordinal responses. I address this open area of research and develop an efficient collapsed Gibbs sampler to estimate such models in univariate and multivariate settings. The paper discusses issues pertaining to model identification and prior sensitivity and provides applications from education and banking.

Work In Progress

The Impact of Bank Resolution Procedures on Risk-taking: Evidence from a Policy Change in the 1980s

Efficient Bayesian Estimation of the Mixed Logit Model with Ivan Jeliazkov and Kai Yoshioka

Bayesian Inference with Synthetic Control Methods with Christoffer Koch and Gary Richardson

Bayesian Joint Modeling with Shape Constraints with Trambak Banerjee

Competitive Effects of Bank and Thrift Rescues: Evidence from Bank-level Portfolios


I have been a TA for the following classes at UC Irvine.


Econometrics (Winter 2016, Spring 2016, Spring 2017, Spring 2018), Probability and Statistics (Summer 2016), Computational Camp (Summer 2016)

Upper Division

Econometrics II (Winter 2017, Winter 2018), Cultural Economics (Summer 2016), Applied Econometrics I (Fall 2016), Applied Econometrics II (Fall 2015), Game Theory (Summer 2015), Managerial Economics (Spring 2015)

Lower Division

Probability and Statistics for Social Sciences (Winter 2015), Probability and Statistics II (Fall 2014)


The latest version of my CV can be accessed here