Information will be updated throughout the summer and fall.

Business Economics

Paul Goldsmith-Pinkham

Household Debt Relief and the Financial Crisis
We use 1.2 million individual credit reports and exogenous variation in state laws to assess the impact of debtor protections during the Great Recession. We find that both bankruptcy homestead exemptions and non-recourse protections helped homeowners reduce their debt. However, while bankruptcy protections raised regional consumption and employment, non-recourse protections lowered both. These contrasting results are explained by higher foreclosure rates and lower regional housing wealth in non-recourse areas leading to lower aggregate consumption, with no similar negative externality in areas protected by bankruptcy exemptions.
Faculty Advisor(s): D. Scharfstein, A. Shleifer, and G. Imbens

Rezwan Haque

Technological Innovation and Productivity in Service Delivery: Evidence from the Adoption of Electronic Medical Records
Information technology (IT) can improve productivity by simultaneously affecting operational efficiency for a given worker and the coordination of different workers. I investigate IT adoption in a service setting by considering the impact of electronic medical records (EMRs) on the length of stay and clinical outcomes of patients in US hospitals. To uncover the distinct impacts of EMRs on operational efficiency and care coordination, I present evidence of heterogeneous effects by patient complexity. I find that EMRs have the largest impact for relatively simple patients. Admission to a hospital with an EMR is associated with a 2% reduction in length of stay and a 9% reduction in thirty-day mortality for such patients. In contrast, there is no statistically significant benefit for complex patients. However, I present three additional results for complex cases. First, patients returning to the same hospital benefit relative to those who previously went to a different hospital, which could be due to easier access to past electronic records. Second, computerized order entry is associated with higher billed charges. Finally, hospitals that have a high share of publicly insured patients, and hence a bigger incentive to curb resource use, achieve a greater reduction in length of stay for complex patients after EMR adoption.
Faculty Advisor(s): D. Cutler, R. Huckman, and O.Hart

Benjamin Hebert

Moral Hazard and the Optimality of Debt
Why are debt securities so common? I show that debt securities minimize the welfare losses from the moral hazards of excessive risk-taking and lax effort. For any security design, the variance of the security payoff is a statistic that summarizes these welfare losses. Debt securities have the least variance, among all limited liability securities with the same expected value. The optimality of debt is exact in my benchmark model, and holds approximately in a wide range of models. I study both static and dynamic security design problems, and show that these two types of problems are equivalent. The models I develop are motivated by moral hazard in mortgage lending, where securitization may have induced lax screening of potential borrowers and lending to excessively risky borrowers. My results also apply to corporate finance and other principal-agent problems.
Faculty Advisor(s): E. Farhi, D. Laibson, D. Scharfstein, J. Campbell, and P. Aghion

James Lee

Measuring Agglomeration: Products, People, and Ideas in US Manufacturing, 1880-1990
I identify local, inter-industry spillovers net of congestion and competition forces in US manufacturing over the twentieth century. I do so along four “Marshallian” industry connectivities: input supplying, output consuming, labor pooling, and ideas collaborating. For identification, I use two major inventions—automobiles in 1904 and semiconductors in 1958—in newly-digitized, city-industry-year level US Census of Manufactures records, 1880-1990. Because the inventions were large, unanticipated demand shocks to supplier industries, the pre-invention supplier share of a city's manufacturing employment is an exogenous measure of the city's invention shock. Hence, a comparison of pre- and post-invention employment between supplier-connected and -unconnected industries, across cities with large versus small pre-invention supplier shares identifies the net connectivity spillovers. In the early twentieth century, net connectivity spillovers were near zero, except for negative net output consuming spillovers. In the late twentieth century, net output consuming spillovers attenuated to zero while net labor pooling spillovers became negative. These results are consistent with falling transportation costs, increased occupational specialization, and reduced worker migration. Together, they point to limited and decreasing local, inter-industry productivity spillovers relative to congestion and competition forces in twentieth century US manufacturing.
Faculty Advisor(s): E. Glaeser, C. Goldin, R. Hornbeck, and W. Kerr

Assaf Romm

An Approximate "Law of One Price" in Random Assignment Games
Assignment games represent a tractable yet versatile model of two-sided markets with transfers. We study the likely properties of the core of randomly generated assignment games. If the joint productivities of every firm and worker are i.i.d bounded random variables, then with high probability all workers are paid roughly equal wages, and all firms make similar profits. This implies that core allocations vary significantly in balanced markets, but that there is core convergence in even slightly unbalanced markets. For the benchmark case of uniform distribution, we provide a tight bound for the workers' share of the surplus under the firm-optimal core allocation. We present simulation results suggesting that the phenomena analyzed appear even in medium-sized markets. Finally, we briefly discuss the effects of unbounded distributions and the ways in which they may affect wage dispersion.
Faculty Advisor(s): A. Roth (Chair), I. Ashlagi, D. Fudenberg, and A. Hassidim

Martin Rotemberg

Competitive Spillovers: Evidence from a Policy Change in India
Abstract forthcoming
Faculty Advisor(s): R. Hornbeck (Chair), M. Kremer, R. Pande, and S. Cole

Ran Shorrer

Consistent Indices
In many economically interesting decision making settings, it is useful to have a complete order over choices that does not refer to the particular preferences of an individual decision maker. I introduce an approach which requires, however, that rankings be consistent with comparisons of preferences. Applications in four settings are introduced: two in risk (the riskiness of gambles and portfolios), time preferences (the delay embedded in investment cashflows) and information acquisition (the appeal of information transactions). In all cases, a unique index is derived, and all indices share several attractive properties. Three of the indices have been introduced elsewhere, based on other approaches, but the index of delay is novel.
Faculty Advisor(s): E. Maskin (Co-chair), A. Roth (Co-chair), and I. Ashlagi

Thomas Wollmann

Trucks Without Bailouts: Endogenous Product Characteristics for Commercial Vehicles
In differentiated product markets, the entry and exit of individual product models--rather than of firms--often serve as the primary equilibrating force. Since market structure changes that lead to high prices also tend to encourage entry, partially offsetting their impact, accurate predictions of welfare and profit changes resulting from a merger or bankruptcy should incorporate this behavior. I develop a model of equilibrium product characteristics in an oligopoly and describe a scheme to set identify the sunk costs of offering them. I apply these methods to a unique dataset of all US commercial vehicle offerings from 1987 to 2012 and assess their importance in the context of the $85B bailout of General Motors and Chrysler. In the case where both firms are liquidated, the entry and exit of product models moderates the welfare losses and markup increases for the most affected consumers and products by over 50%. There is, however, little impact on the median buyer and model in the market. I also consider counterfactuals where these firms are acquired by a rival, which yields similar results.
Faculty Advisor(s): A.Pakes (Chair), J. Alcacer, G. Lewis, and D. Yao

David Yang

Does the Tail Wag the Dog? How Options Affect Stock Price Dynamics (with Fan Zhang)
We empirically document how options affect stock price dynamics. In contrast, in classical asset pricing, options “derive” their value from the underlying stock. Our mechanism is similar to portfolio insurance, which some argue exacerbated the 1987 stock market crash. In particular, “hedging demand” by option writers creates an upward sloping demand curve: when the stock price rises, option writers must buy more stock to remain hedged (and vice versa). This hedging demand increases the autocorrelation of stock returns. We study this effect for options on individual stocks. As we move from the lowest to highest quintile of log hedging demand, the average return autocorrelation increases from -4.9% to -1.5%. In a portfolio sorting strategy, the high-minus-low-quintile return spread has an alpha of 0.08% daily (22% annualized) before transactions costs. To establish causality, we use a novel instrument for hedging demand: distance to the nearest round number price.
Faculty Advisor(s): J.Campbell, L. Summers, and R. Greenwood


Elizabeth Altman

Platform and Ecosystem Transitions: Organizational and Strategic Implications
“Joining an Ecosystem: Organizational and Strategic Implications”
With the increasing worldwide popularity of products that include open interfaces such as smartphones, tablets, and personal computers, individuals are becoming increasingly comfortable with complementary products where a product becomes more useful when it is combined with other products, such as applications and accessories. Organizational theory and strategy research on platforms and ecosystems address these dynamics, but tend to center on organizations competing at the core of ecosystems. This paper focuses on organizations at the edges of these systems that join ecosystems and must follow rules determined by others. Based on a three year qualitative longitudinal inductive field-based study, this research investigates organizational and strategic changes experienced by a firm as it joins an ecosystem and strives to balance maintaining its independence and growth aspirations with its need to operate within an ecosystem. I find organizational identity implications driven by asymmetries in power, communication challenges, and loss of control. As the organization experiences increased interdependence, it balances maintaining independence and growth aspirations with ecosystem compliance. This paper contributes to research on organizational identity, resource dependence, and asymmetric inter-organizational relationships, and also complements burgeoning strategy and organizational research on ecosystems and multi-sided platforms.
Faculty Advisor(s): M. TushmanA. Hagiu, M. Tripsas, and K. Kellogg


Lalin Anik

Contingent Match Incentives Increase Donations
We propose a new means by which non-profits can induce donors to give today and commit to giving in the future: contingent match incentives, in which matching is made contingent on the percentage of others who give (e.g., “if X% of others give, we will match all donations”). A field experiment shows that a 75% contingent match (where matches “kick in” only if 75% of others donate) is most effective in increasing commitment to recurring donations. An online experiment reveals that the 75% contingent match drives commitment to recurring donations because it simultaneously provides social proof yet offers a low enough target that it remains plausible that the match will occur. A final online experiment demonstrates that the effectiveness of the 75% contingent match extends to one-time donations. We discuss the practical and theoretical implications of contingent matches for managers and academics.
Faculty Advisor(s): M. Norton (Chair), J. Gourville, T. Steenburgh, and E. Dunn

Silvia Bellezza

Symbolic Consumption: How Consumers Use Products, Brands, and time to Express Identity and Signal Status
My dissertation is composed of three papers on symbolic consumption–how individuals engage in consumption practices and spend time to express who they are and to signal status to others. In the first paper (“Brand Tourists: How Non–Core Users Enhance the Brand Image by Eliciting Pride,” forthcoming in the Journal of Consumer Research), I demonstrate the positive role of brand tourists, non-core users of exclusive brands perceived as fans of the brand community. In the second paper (“The Red Sneakers Effect: Inferring Status and Competence from Signals of Nonconformity,” published in the Journal of Consumer Research), I explore the conditions under which nonconforming behaviors, such as wearing red sneakers in a professional context or entering a luxury boutique wearing gym clothing, lead to inferences of higher status and competence in the eyes of others. In the third paper (“Conspicuous Consumption of Time: When Busyness at Work and Lack of Leisure Time Become a Status Symbol,” under review at the Journal of Consumer Research), I further extend my research on subtle signals of status by uncovering the role of busyness at work and lack of leisure time as a status symbol.
Faculty Advisor(s): A. Keinan (Chair), F. Gino, and J. Gourville

Organizational Behavior

Faaiza Rashid

Mutual Accountability in Teams
Using qualitative and quantitative methods, I study the dynamics of mutual accountability in teams. Understanding how and under what conditions mutual accountability occurs is an important part of explaining why some teams live up to their potential and perform better than others. Many teams, especially in fast-paced knowledge-intensive environments, face interdependent tasks with unscripted responsibilities. The centrality of this challenge to the team process notwithstanding, theories of how team members hold one another accountable for accomplishing interdependent work are underdeveloped. In my dissertation, I integrate theory and research on accountability and teams to introduce the construct of team mutual accountability, a shared expectation among team members that they are authorized to evaluate one another's progress on the team's task. In a field study of 50 teams from five knowledge-intensive organizations, I collected and analyzed data from interviews and surveys to examine whether mutual accountability varies across teams, under what conditions mutual accountability occurs in teams, and what are the effects of mutual accountability on the team process and performance. Results show that, controlling for externally enforced top-down accountability on a team, team mutual accountability is positively associated with the effectiveness of team process and team performance. Team mutual accountability is facilitated by a context that rewards and supports teamwork and a shared understanding among team members of the team's task, plan of action, and one another's roles and responsibilities. The findings of this research advance theory and research on teams and accountability.
Faculty Advisor(s): A. Edmondson (Chair), R. Ely, and F.Garip

Ting Zhang

The Unexpected Benefits of Rediscovery
My dissertation focuses on the psychology of rediscovery, which describes the process of revisiting past experiences. In Chapter 1 (“A ‘present’ for the future: The unexpected value of rediscovery,” forthcoming in Psychological Science), I demonstrate that people underestimate the value of rediscovering today’s ordinary moments in the future. Consequently, individuals make time-inconsistent choices: they forgo opportunities to document the present, but then prefer to rediscover those moments in the future. Although people generally overlook the value of rediscovering the present in the future, the act of rediscovery can be a simple, but powerful intervention that shapes how individuals interact with one another and engage in their current work. Chapter 2 focuses on how rediscovery of the past helps experts overcome the curse of knowledge and better relate to novices in the context of medicine and music. I show that rediscovering past experiences helps experts give more actionable advice to novices compared to merely reflecting about the experience of being a novice. Finally, Chapter 3 investigates how rediscovery of past work-related events helps individuals realize the progress they have made and find meaning in their jobs, better equipping them to face upcoming work-related challenges.
Faculty Advisor(s): F. Gino (Chair), M. NortonM. Bazerman, and J. Margolis

Pavel Zhelyazkov

Presence of A Tie or Content of the Tie: Venture Capital Syndication Networks and Limited Partner Investment Decisions
Network theory has long assumed interorganizational relationships act as pipes transmitting information across the market, but have rarely considered the effect of the content of such information on organizations’ behavior. In particular, scholars have attributed the well documented phenomenon of triadic closure – the tendency of actors to form direct ties with those that they are indirectly connected to through shared partners – to the flow of referrals across the indirect ties, without considering whether such referrals are positive or negative. I argue that the effect of indirect ties will be contingent on the performance of the exchange between the shared partner and the indirectly connected actors: success will increase the chance of positive referrals and positive effects of indirect ties on direct tie formation, while failure will engender negative referrals and negative effects of indirect ties on tie formation. Furthermore, the reputation of the shared partner will magnify the effects of both positive and negative referrals. Finally, actors will reduce their reliance on referrals if they have access to other sources of public or private information about the prospective counterparties. I find support for these hypotheses with a longitudinal dataset of the investment decisions of Limited Partners (LP) investing in Venture Capital (VC) firms.
Faculty Advisor(s): R. Gulati, P.Marsden, M. Torfason, and P. Gompers


Andrea Hugill

Political Risk, Knowledge, and Investment Strategy
Using investment data from mobile telecom operators from 2000-2010, I examine whether firms with heterogeneous access to knowledge had different investment strategies in politically risky markets. I find that firms with access to local knowledge, gained through geographic proximity to the local market, were more likely to decrease investment as political risk rose to the highest levels. In contrast, firms with access to experiential knowledge, gained from operating in a range of institutional settings across markets, were more likely to increase investment as political risk rose to the highest levels. These results offer two main contributions: First, I present analysis of market-based approaches firms take when confronting political risk. Second, I document distinct strategies associated with local and experiential knowledge for navigating political risk. While locals and foreigners are understood to have knowledge-based competitive advantages in varied institutional settings, I show that, in institutional settings characterized by political risk, these distinct forms of knowledge also have unique strategic implications.
Faculty Advisor(s): J. Siegel (Chair), J. Alcacer, and D. Yao

Technology & Operations Management

Sen Chai

Temporary Colocation and Collaborative Discovery
The lack of empirical evidence on the impact of conferences on participants has fueled a heated debate. On the one hand, researchers are advised to attend conferences to further their careers, but at the same time there are obvious trade-offs of diverted funding and potential productivity loss while away from the bench. I investigate how temporarily colocating at conferences affects attendees’ research trajectory, in terms of collaborations and citations. I use difference-in-differences regressions on a sample of attendees from Gordon Research Conferences and most similar matched researchers, and several different cuts of the data to address endogeneity of better researchers selected to present, existing co-authors attending together and choosing to go to a conference. My results suggest that even after a transitory period being colocated, long-term collaborations between conference attendees increase with especially strong effects for those who have never published together beforehand. Conditional on collaborative ties forming, I find collaborative outputs between conference attendees draw more from the knowledge space of the conference and are also more highly cited. Conferences also enable attendees who have never been cited by other attendees to showcase their research as evidenced by increases in within-attendee citations. Given the cumulative nature of research, these findings imply that over time conferences can have a significant impact in steering the research path of attendees, from the works that they cite and build upon to the colleagues with whom they choose to collaborate.
Faculty Advisor(s): L. Fleming (Chair), G. PisanoV. Sato, and and F. Murray

Anil Doshi

The Impact of High Performance Outliers on Two-Sided Platforms: Evidence from Crowdfunding
How do differences among users on one side of a two-sided platform affect the platform's growth and liquidity? I focus on the arrival of high-performing sellers, or stars, and subsequent seller entry and buyer transactions. In the context of crowdfunding, I find that the arrival of a star on the dominant platform results in a decline in seller entry and transactions, relative to the competing platform. Within a platform, I find evidence of an increase in entry and transactions for sellers that are similar to the star on the dominant platform. The impact of stars is dependent upon how stars are defined and the characteristics of the star. The results in this paper suggest in addition to pricing and platform structure, competing platforms may selectively focus on attracting users with high performance potential to create competitive advantage.
Faculty Advisor(s): S.Stern, L. Fleming, D. Yao, and F. Zhu

Frank Nagle

The Digital Commons: Tragedy or Opportunity? The Effect of Crowdsourced Digital Goods on Innovation and Economic Growth
In the digital world, public and crowdsourced goods are essentially infinite. Although this solves the classic tragedy of the commons problem, the widespread digital availability of goods that were once scarce results in a destruction of value as measured by traditional means. In this dissertation, I examine this phenomenon to better understand how digitization is weakening the boundaries of the firm while enabling innovation and economic growth. In paper 1 (w/ Shane Greenstein) we explore the impact of "digital dark matter", digital goods and services that are non-pecuniary and effectively limitless inputs into production, on GDP measurement. We find that one such good, the Apache web server, leads to an underestimation of GDP by upwards of $12 billion. In paper 2 (my job market paper), I empirically measure the impact of non-pecuniary crowdsourced digital goods on firm productivity. Using data from a survey of technology use at nearly 2,000 firms over 10 years, I find that a 1% increase in the amount of non-pecuniary open source software used by a firm leads to a .075% increase in productivity. This translates to a $1.38 million increase in production output for the average firm in my sample. I use inverse probability weighting, instrumental variables, firm-fixed effects, and data on managerial quality from the World Management Survey to add support to a causal interpretation of these results. Paper 3 (w/ Elizabeth Altman and Michael Tushman) presents a conceptual argument for why technological progress and reductions in information costs are leading firms to increasingly engage with external digital communities in ways that alter traditional theories of how firms organize and innovate. Finally, in paper 4, I empirically examine how firm participation in such communities can enhance the absorptive capacity of the firm using a dataset on firm contributions to open source software projects.
Faculty Advisor(s): S.Greenstein (Co-chair), M. Iansiti (Co-Chair), C. BaldwinK. Lakhani, and F. Zhu

Tina Tang

Strategy in Innovative Industries
My thesis explores the firm’s role in shaping the adoption of new products in industries with frequent innovation and free entry. It consists of three essays: a microeconomic model of adoption under network effects, a study of geographic entry strategies in the adoption of, and an empirical model of first mover advantage in markets for exchange-traded funds (ETFs).
Faculty Advisor(s): R. Casadesus-MasanellM. IansitiM. Luca, and H. Luo