Latest Research Publications

In the highly competitive mobile market, third-party vendors located outside the purview of hosting mobile platforms are becoming major suppliers of functional tool kits for mobile app development and innovation. Mobile app developers, however, face the uncertainty of whether and how to use third-party software development kits (SDKs) from these external vendors to create more appealing and engaging mobile apps. This study examines the extent to and conditions under which third-party SDK utilization affects mobile app market performance. Drawing on the platform ecosystem literature and boundary object theory, we contextualize the boundary-spanning practice in mobile app development as the extent to which developers utilize third-party SDKs and theorize the performance impact of third-party SDKs. Moreover, the boundary-spanning perspective leads us to examine how the performance impact of third-party SDKs varies across tool types versus platform types, the evolution of platform boundaries, and levels of app developers’ platform-specific experience. By conducting difference-in-differences-style analyses on a longitudinal data set of 335,958 multihoming mobile apps released on Apple App Store and Google Play Store, our study reveals that utilizing more third-party SDKs is positively associated with daily active users of mobile apps. This positive impact is limited to tool-type third-party SDKs, however, and is attenuated by platform updates and app developers’ platform-specific experience. This study contributes to the platform-based software innovation and platform governance literature and provides managerial implications for app developers, platform managers, and third-party SDK providers.

We study competition between two firms that personalize product offerings to consumers. Firms have private, imperfect signals of each consumer’s ideal location and offer each consumer a different product without observing the competitor’s product offering. We characterize the equilibrium personalization strategy and examine how the accuracies of firms’ signals affect equilibrium strategy, profits, and consumer welfare. A firm generally charges a higher price for a more niche product and profits more from niche consumers unless its prediction accuracy is sufficiently lower than its competitor’s. When both firms have the same industry-level prediction accuracy, an increase in accuracy initially relaxes but later intensifies price competition for niche consumers, having the opposite effect on mainstream consumers. Interestingly, equilibrium profits also have an inverse-U shape in the prediction accuracy. A higher accuracy can also decrease welfare for mainstream consumers. When firms can endogenously invest in prediction accuracy, firms have incentives to overinvest in equilibrium, resulting in a prisoner’s dilemma. Privacy regulations that reduce predictive accuracy, including industry self-regulation, could improve profits and hurt consumer welfare by relaxing price competition. Our results remain robust under consumer search. The paper also discusses what happens if firms charge uniform pricing, if consumers’ ideal locations are distributed on the Salop circle, or if firms receive common signals, highlighting price discrimination between mainstream and niche consumers as the key driver of results.

Lack of comovement between consumption differentials and real exchange rates is a traditional indicator of a disconnect of foreign exchange markets from economic fundamentals. We present novel empirical evidence for the disconnect between the volatilities, as opposed to the levels, of these variables. The volatility correlations are below one, but they are larger than the level correlations. We discuss the economics of volatility disconnect anomaly in settings with complete and incomplete markets and provide an explanation of our empirical findings based on international risk sharing of expected growth and volatility news shocks.

Domestic outsourcing is known to reduce worker wages, but its effect on employment security — a key dimension of job quality — has not been studied. Using Brazil’s comprehensive employee–employer linked data, we find that outsourcing reduces exit from formal employment among cleaners and security guards during their first few years of tenure. The observed reduction in employment hazard is larger in cities with greater volatility in labor demand. The reduction is not attributable to differences in worker characteristics or differential exposure to local labor market shocks. The estimates suggest that outsourcing had larger positive effects on the net present value of worker earnings than implied by wage differentials alone. The patterns are consistent with a search-theoretic model in which outsourcing eases reassignment across firms.

Over the past decades, research on gender bias in leader evaluations has proliferated across multiple disciplines, significantly expanding contexts, outcomes, and theoretical perspectives examined. Despite these valuable contributions, the literature remains fragmented in explaining the persistent variability in how and why gender bias manifests, from severe penalties against women leaders in certain contexts to evaluative advantages in others. To resolve these discrepancies, we shift the focus from leaders to the motivated processes driving observer evaluations. We begin with an integrative review of research, revealing that observers, ranging from supervisors and subordinates to clients and investors, are not conduits of stereotypes but active evaluators whose motives shape how they selectively appraise women leaders. Drawing on motivated cognition theory, we develop a novel motive-driven process model that identifies three core directional motives: identity protection, value alignment, and resource dependence. Using this model, we integrate the literature by highlighting individual-level and contextual antecedents of each motive and explicating how motives drive distinct selective appraisal processes. By unpacking “why” and “how” observers evaluate women leaders through motivated processes, our model also offers targeted interventions that address observer motives rather than changing women’s behaviors, underscoring a pressing need to engage various stakeholders in addressing gender bias in leader evaluations.

In this paper, we characterize a forecasting model where forecasters cannot perfectly distinguish between the two persistent components (trends and cycles) in a dynamic setting. In this model, forecasters jointly update their beliefs about the two components: noisy information about one component is used to update beliefs about the other component. We present diagnostic empirical facts on forecasting behaviors and show that these facts are consistent with our model’s predictions while contradicting those of existing models in the expectation formation literature. To validate our model, we exploit the Federal Reserve’s 2012 adoption of explicit inflation targeting as a policy shock. Structural estimation reveals that this policy change altered the underlying data-generation process, and the corresponding changes in forecasting behavior indeed align with our model’s predictions. Finally, we revisit the standard Forecast Error-Forecast Revision regression approach in this literature. We examine its robustness within our enriched framework and reveal that trend-cycle confusion can interact with behavioral bias and generate horizon-dependent overreaction patterns documented in empirical studies.

A genealogical training process, in which senior (advisor) scientists mentor and train junior (advisee) scientists is one of the core organizational features of modern science. In this paper, we examine a key question faced by all junior scientists during their training: What impact does an advisee’s research agenda overlap with his or her advisor have on the advisee’s career-relevant performance outcomes? To answer this question, we constructed a novel, bibliometric-record-based data set on 11,289 U.S. biomedical scientists (advisees) who were trained in 5,632 principal investigator advisors’ labs between 1985 and 2009. We examined the relationship between advisor–advisee research overlap and an array of performance outcomes for emerging scientists, revealing a consistently positive relationship between high advisor–advisee research overlap and the junior scientist’s early-career funding outcomes. We further provide evidence that this positive relationship rests upon enhanced tacit knowledge transfer, as well as providing suggestive evidence for the boundary conditions of an intellectual independence imperative and potential competition between advisors and advisees. Taken together, these findings provide a more complete understanding of how advisor–advisee relationships shape new scientists’ performance during their early careers.

We study dynamic contracts that incentivize an agent to exert effort to increase the arrival rate of a Poisson breakthrough, where both the effort cost and the effort level at any time are the agent’s private information. Optimally, the principal offers a menu of contracts, each tailored to an agent type (with a different effort cost), specifying an initial payment, a contract deadline, and a payment-upon-arrival process over time. We first fully characterize the optimal contract menu in a two-type setting, where the agent is either a good (low-cost) or bad (high-cost) type. Specifically, the principal should hire the agent only if the breakthrough revenue exceeds a threshold. Above this threshold, if the bad agent’s cost is higher than another threshold, it is optimal to motivate only the good type to exert effort. The principal offers the good type a simple linear contract in which the payment-upon-arrival declines linearly over time until the deadline, whereas the bad type receives an initial payment and leaves immediately. The linear contract provides just enough incentive for the good agent to work. If the bad agent’s cost falls below the threshold, it becomes optimal to also motivate the bad agent to work using a linear contract, whereas offering the good agent a one-switch contract. The one-switch contract extends the linear form by allowing the payment-upon-arrival to take a single downward jump at a specific time before the deadline. The optimal contract structure extends to multiple-type cases, in which the one-switch contract becomes a multiple-switch contract. To obtain the entire menu of contracts, one only needs to solve a sequence of linear optimization problems together with a bisectional line-search, which is fast to compute and easy to interpret and implement.

Moment-based distributionally robust optimization (DRO) provides an optimization framework to integrate statistical information with traditional optimization approaches. Under this framework, one assumes that the underlying joint distribution of random parameters runs in a distributional ambiguity set constructed by moment information and makes decisions against the worst-case distribution within the set. Although most moment-based DRO problems can be reformulated as semidefinite programming (SDP) problems that can be solved in polynomial time, solving high-dimensional SDPs is still time-consuming. Unlike existing approximation approaches that first reduce the dimensionality of random parameters and then solve the approximated SDPs, we propose an optimized dimensionality reduction (ODR) approach by integrating the dimensionality reduction of random parameters with the subsequent optimization problems. Such integration enables two outer and one inner approximations of the original problem, all of which are low-dimensional SDPs that can be solved efficiently, providing two lower bounds and one upper bound correspondingly. More importantly, these approximations can theoretically achieve the optimal value of the original high-dimensional SDPs. As these approximations are nonconvex SDPs, we develop modified alternating direction method of multipliers algorithms to solve them efficiently. We demonstrate the effectiveness of our proposed ODR approach and algorithm in solving multiproduct newsvendor and production-transportation problems. Numerical results show significant advantages of our approach regarding computational time and solution quality over the three best possible benchmark approaches. Our approach can obtain an optimal or near-optimal (mostly within 0.1%) solution and reduce the computational time by up to three orders of magnitude.