Sumedh Ambokar

Sumedh Ambokar
Job Market Paper

Mortgage Search Heterogeneity, Refinancing Decisions and Monetary Policy Transmission to Consumption (with Kian Samaee)

In the US, half of all mortgage borrowers consider one lender at origination. This paper investigates how heterogeneity in mortgage search affects refinancing decisions, the distribution of home equity and mortgage rates. These affect the consumption response to a monetary policy shock at the aggregate as well as distributional level. We build a general equilibrium model of the mortgage market with two types of mortgage borrowers: they get either one quote or two quotes at refinancing. A lender infers the expected search behavior from the relative mass of the two types at any observable current rate and home equity. If the refinancer is more likely to get one quote, then lenders have more market power. As a result, they offer higher rates to both types, reducing the likelihood that they refinance. So, even a borrower with high current rate may decide not to refinance as she is offered high rates because of being inferred as more likely to get one quote. In steady state, those who get two quotes get lower rates sooner and hence accumulate more home equity. As such, in response to a 25 basis points expansionary monetary shock, the percentage increase in consumption of those who get two quotes is 1.47 times that of those who get one quote. In a counterfactual economy with more mortgage search intensity, an explicit goal of the CFPB, percentage increase in borrower consumption in response to the same shock is 1.52 times that in the benchmark model. Thus, more mortgage search results in significantly higher consumption response to expansionary monetary policy.

Other Research

Inaction, Search Costs and Market Power in the US Mortgage Market (with Kian Samaee)

Many US mortgage borrowers do not refinance despite seemingly having financial incentives to do so. We explore the role of search costs in explaining this inaction, focusing on the 2009-2015 period when mortgage rates significantly declined. We estimate a (dynamic) discrete choice model of refinancing and search decisions using a proprietary panel data set, which include detailed information on mortgage contracts, borrower creditworthiness and search intensity (number of mortgage inquiries), and the sequence of refinancing decisions. We find that search costs significantly inhibit refinancing through two channels: While larger search costs directly increase the cost of refinancing, they also indirectly increase loan originators’ market power and raise the offered mortgage rates. We find that the indirect market power effect dominates. We also estimate extensive heterogeneity in mortgage approval rates across the borrower credit quality distribution, which affect their search intensity and refinancing activity. We apply our model to counterfactual policy exercises intended to encourage refinancing activity by removing credit quality from lender approval decisions.  We find that these policies are not effective because most borrowers have such high estimated search costs that even higher approval rates do not significantly change their search intensity. We conclude that removing refinancing barriers from the credit supply side is unlikely to improve refinancing activity given borrower lack of mortgage shopping.

Effect of Coalition Governments on Public Investment

Based on a panel data of 80 countries, I find that during an election cycle, public investment is a concave function of the months to election. Also, the post-election increase in public investment is smaller if the share of the coalition partners in the government is higher and that a government perceived to be corrupt does less public investment. I write a model of coalition governments where parties in government divert a fraction of the imperfectly observed public investment for private benefit and the smaller party in government can call an early election hoping to become the bigger party post-election. This model adds endogenous early elections to models of political competition. Coalition governments invest less to signal honesty and thus avoid an early election. Strong governments invest less as election approaches to signal honesty whereas weak governments invest more as election approaches as the incentive of smaller party to call an early election reduces as the scheduled election is nearer; thus public investment is concave during an election cycle. For optimal growth-inducing public investment, winner-takes-all systems are better than proportional systems to reduce these distortions.

Teaching Experience


Macroeconomic Theory I, University of Pennsylvania, Teaching Assistant for Prof. Jose-Victor Rios-Rull (Spring 2017), for Prof. Dirk Kruger (Spring 2016)

Network Security, Indian Institute of Technology, Bombay, Teaching Assistant for Prof. Bernard Menezes (Fall 2009)


Fiscal and Monetary Policy, University of Pennsylvania, Teaching Assistant for Prof. Harold Cole (Spring 2018)

Introduction to Microeconomics and Macroeconomics, University of Pennsylvania, Teaching Assistant for Prof. Saka (Fall 2018 (Head TA), Fall 2017), Prof. Saka and Prof. Duchene (Fall 2016, Fall 2015)

Cryptography and Network Security, Indian Institute of Technology, Bombay, Teaching Assistant for Prof. Bernard Menezes (Spring 2010)

Other Information


Professional Experience

AQR Capital Management, PhD Research Intern (2019)

Goldman Sachs, Quantitative Strategist (2014)

Oracle, Member Technical Staff (2010-12)


Computational Skills: Python, Matlab, Stata, R, C++, Java, SQL, C


Languages: English (Fluent), Hindi (Native), Marathi (Native)



“An Architecture for regulatory compliant database management”, Proceedings of the 2009 IEEE International Conference on Data Engineering, IEEE Computer Society, 2009. (with Soumyadeb Mitra, Marianne Winslett, Richard T. Snodgrass and Shashank Yaduvanshi)


Immigration Status

Visa: F-1 (Eligible for 24-month STEM OPT extension)

Citizenship: India


Macroeconomics, Mortgage Markets, Monetary Economics


Department of Economics
University of Pennsylvania
133 South 36th Street, Office 548
Philadelphia, PA 19104


+1 (917) 624-7663


Download CV


Harold L. Cole


Professor Harold Cole (Advisor)

Department of Economics

University of Pennsylvania

+1 (215) 898-7788


Professor Aviv Nevo

Department of Economics

University of Pennsylvania

+1 (215) 898-0499


Professor Benjamin Keys

Wharton Real Estate Department

University of Pennsylvania

+1 (215) 746-1253


Job Market Candidate Status
I am on the job market and I will be available for interviews at the 2020 ASSA Annual Meeting in San Diego