Interests

I find it fascinating to study decentralized decision-making and its collective impact. When institutions, organizations, or individuals make decisions that affect the choices of others, the final results can often be complex and lead to unintended consequences. I use mathematical models of markets and games to predict outcomes in these situations and to develop principles to guide predicted outcomes. My research uses analytically rigorous arguments and quantitative methods to derive conclusions that are more likely to survive the test of time. A brief summary of my research classified into broad categories is included below. Published papers and their abstracts are available here. 

I have studied in detail some classes of decentralized decision-making in socio-economic situations characterized by monotone interdependent incentives, which include situations in which participants have incentives to move in particular directions, either coordinated moves, or opposing moves, or a combination of the two. Previously existing results dealt mostly with the case of coordinated moves (or strategic complements). As other cases were hard to prove directly, researchers frequently resorted to transform non-complements cases into those with complements and apply the results for complements. This approach had limited practical success and frequently changed the natural interpretation of monotonicity in the other cases, making the applications counter-intuitive. Nevertheless, it was a well-established belief that the main problem in studying more general cases was how to transform them into strategic complements. 

Over a period of 10-15 years, my research has overturned this conventional wisdom, transformed the understanding of the fundamental forces underlying the new cases, developed new tools to analyze directly interdependent interactions involving strategic substitutes and heterogeneity, and expanded the understanding of related influential literatures like directional optimization, global games, and dynamic games. My papers on these topics include Roy and Sabarwal (2008, 2010, 2012), Monaco and Sabarwal (2016), Barthel and Sabarwal (2018), Hoffmann and Sabarwal (2015, 2019a, 2019b), Feng and Sabarwal (2020), Barthel, Hoffmann, and Sabarwal (2022), and Sabarwal and Vu (2020)). Based on the success of this research, I was awarded a contract to write a research monograph on monotone games (Sabarwal (2021)). 

My research on financial markets sheds light on different aspects of the design, pricing, and performance of asset-backed securities, irreversible investments, and limited liability debt contracts in financial markets. Theoretical underpinnings of these ideas are formulated in the early study of chain reactions of personal bankruptcy as an equilibrium phenomenon in anonymous asset-backed securities markets with limited liability consumer debt contracts in the setting of multi-period, multi-asset, multi-good, multi-agent, incomplete and competitive markets in Sabarwal (2003), the first paper with this level of generality. That paper also showed how allowing for bankruptcy can increase market efficiency, improving borrower welfare and lender welfare by endogenously creating new assets with better risk-sharing opportunities. These ideas paved the way for the first academic paper on the drivers of default and prepayment on subprime auto loans using hazard rate modeling and data from asset-backed securities in Heitfield and Sabarwal (2004). Common structures used in the design of asset-backed securities and their associated risks are discussed in Sabarwal (2006). 

The equilibrium effect of limited liability debt contracts on the NPV investment threshold for irreversible investments in real options theory is studied in Sabarwal (2005). This presented the first derivation of the NPV threshold as an equilibrium phenomenon in models of irreversible investment with a competitive lending sector. 

Studying antitrust aspects of investment banking, Kulkarni and Sabarwal (2007) show that the high and stable spreads charged to bring moderate-sized IPOs to market are hard to explain using the standard argument that investment banks provide differentiated services to firms in different industries. 

I have been intrigued by efficiency-driven mergers and acquisitions and studied some properties of Upward Pricing Pressure (UPP), a new tool that is being used increasingly widely worldwide in the analysis of horizontal mergers. UPP is tractable, easy to implement, uses less information than other standard measures, and is derived from the existing theory of oligopoly. Dutra and Sabarwal (2020) show that the accuracy of UPP as a tool in merger analysis is enhanced greatly by inclusion of merger-specific cost efficiencies directly in the computation. Our formulation of UPP may be an excellent and cheaper alternative to full-merger simulations, which are considerably more expensive to implement. This has the potential to impact antitrust policy worldwide. 

I am excited to develop a new research line at the intersection of economics, computer science, and game theory. There is an explosion of interest in artificial intelligence and neural networks, based on their phenomenal success in solving complex problems in automation, classification, and control. These ideas are based on application of stochastic adaptive control from mathematics and operations research. There is a related field in game theory on stochastic games. Both stochastic games and stochastic control have applications in broad swaths of knowledge, including business, engineering, mathematics, natural sciences, and social sciences. Both fields are technically complex in different ways but share some common features. Presently, there is limited interaction between these fields, partly due to their respective complexity. As part of the Keeler Intra-University Professorship award, I devoted Fall 2020 to study both fields and explore research ideas at the intersection of these areas. This is expected to open new research lines with many applications and opportunities.

A new paper in this area is Higgins and Sabarwal (2023). We study proliferation of an action in binary action network coordination games that are generalized to include global effects. This captures important aspects of proliferation of a particular action or narrative in online social networks, providing a basis to understand their impact on societal outcomes. Our model naturally captures complementarities among starting sets, network resilience, and global effects, and highlights interdependence in channels through which contagion spreads. We present new, natural, and computationally tractable algorithms to define and compute equilibrium objects that facilitate the general study of contagion in networks and prove their theoretical properties. Our algorithms are easy to implement and help to quantify relationships previously inaccessible due to computational intractability. Using these algorithms, we study the spread of contagion in scale-free networks with 1,000 players using millions of Monte Carlo simulations. Our analysis provides quantitative and qualitative insight into the design of policies to control or spread contagion in networks. The scope of application is enlarged given the many other situations across different fields that may be modeled using this framework.

My research in applied microeconomics focuses on decentralized market outcomes from an empirical standpoint, for example, in personal bankruptcy filing and limited liability debt contracts. Earlier research concluded that people file for bankruptcy more due to financial benefit from filing and not due to adverse events such as medical problem, unemployment, or divorce. The earlier conclusion did not account for the financial impact of adverse events on the financial benefit from filing for bankruptcy. Accounting for this reverses the earlier conclusion, and shows that consumers are not gaming the institution of bankruptcy in a systematic manner (see Zhang, Sabarwal, and Gan (2015)). The theoretical underpinnings of these ideas are based on the study of chain reactions of bankruptcy in anonymous asset-backed securities markets in Sabarwal (2003). Factors driving defaults and prepayments on subprime auto loans are studied in Heitfield and Sabarwal (2004). 

A problem in personal bankruptcy research has been lack of publicly available data. Only a few hundred cases are found in publicly available datasets, as compared to more than 40 million bankruptcy filings since 1898. Lenders have some data but it is not representative and is unavailable for academic research. With external funding from the National Science Foundation (2014-17, $176,061) and the Alfred P. Sloan Foundation (2012-13, $48,160), I led research teams to collect bankruptcy data from historical court records at the Kansas City office of the National Archives of the United States. Altogether, the research team took about 525,000 photographs of documents from more than 25,000 cases filed in 24 federal district courts in 18 states over different periods spanning more than a century (1898-2002). These projects funded 4 graduate students and 17 undergraduate students at KU to gain first-hand research experience in original source data collection, a rare and increasingly sought skill in economics.