We examine how the information in public signals (e.g., earnings, analysts’ forecasts) affects firms’ voluntary disclosures. We consider levered firms and information structures such that investors update their beliefs about both the mean and variance of the firm’s cash flows. We predict that firms that, conditional on the public signal, have a higher default risk are less likely to disclose. For firms with low ex ante default risk, larger positive surprises following the public signal initially increase the disclosure likelihood because they indicate a higher mean of cash flows. However, too large positive surprises may reduce the disclosure likelihood of these firms, because the larger surprise is also indicative of higher variance and, thus, higher probability of default. Symmetrically, larger negative surprises may increase disclosure for firms with high ex ante default risk, despite the more pessimistic inference about the mean, because the higher variance makes it more likely that these firms avoid default. Extending our analysis to multiple analysts’ forecasts, we predict that not only consensus, but also dispersion affect firms’ disclosure likelihood. Consensus conveys information about the mean and variance of cash flows, whereas dispersion is informative about the precision of analysts’ signals, which in turn affects the information content of forecasts. This paper was accepted by Suraj Srinivasan, accounting. Funding: D. Lee-Lo gratefully acknowledges financial support from the Accounting Research Center at Kellogg.

This paper examines the consequences of tightening board fiduciary duty rules in a setting that combines agency frictions and strategic communication. A firm needs to tailor an investment decision (e.g., mergers and acquisitions) to the state of the world. The CEO is privately informed about the state but is an empire builder. The board can learn about the firm’s state through exerting a costly information gathering effort or through communication. In equilibrium, the board values communication more than the shareholders do. By granting the CEO a larger equity stake, the board fosters communication, and this reduces its need for costly information acquisition. Anticipating this, rational shareholders can adjust the board’s equity stake to boost its effort incentive. Surprisingly, the benefit to shareholders from stricter fiduciary duty rules is nonmonotonic in CEO agency conflicts and can diminish as the severity of agency conflict increases. Moreover, welfare analysis shows that strengthening fiduciary duty rules can weaken board effort incentives and reduce total equity value and overall welfare. Paradoxically, these efficiency losses occur when CEO agency conflicts are severe—the very situations in which stringent fiduciary duty rules are typically introduced to address such concerns. These findings suggest unintended consequences of strengthening board fiduciary duty rules and highlight the importance of taking boards as strategic, self-interested entities. This paper was accepted by Prof. Ranjani Krishnan, accounting.

We examine the relation between corporate tax planning and firm-level productivity. Using a sample of U.S. public firms from 1993 to 2017, we find that tax planning is positively associated with productivity. Our findings further indicate that tax planning contributes to growth in production input factors: capital and labor. We also test the underlying mechanisms, documenting that the observed association between productivity and tax planning is more pronounced in financially constrained and knowledge capital–intensive firms. We exploit the initiation of new banking relationships and the introduction of Check-the-Box regulations as sources of tax planning variation to facilitate difference-in-differences analyses. The empirical evidence holds consistently across a wide array of robustness tests. Taken together, the results enhance the understanding on the economic implications of tax planning and provide insights for informed tax policy debates. This paper was accepted by Suraj Srinivasan, accounting. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2022.03569 .

Choice models, which specify the consumers’ choice probabilities of options over a given choice set, are widely studied and applied in many fields. We propose a temporal tree representation of choice that covers all rational choice models. Compared with existing structural choice models, the tree representation exhibits two major advantages that address the key challenges of model identification. First, all rational choice models have a tree representation, and a subclass of tree representation (with set-dependent branching) has a one-to-one correspondence with the rational choice models. This bridges the gap in the existing characterizations of structural models, which are unidentifiable, impose uninterpretable conditions, or do not cover the entire space of rational choice models. Second, the tree representation allows for the flexibility of systematically specifying the choice model structure based on available knowledge and data. In particular, the number of parameters needed to specify a tree representation can be primarily determined by the sufficient knowledge level, which corresponds to a specific layer of the tree branching. The sufficient knowledge level can be empirically determined based on the amount of available data, which, in turn, determines the number of parameters needed for model estimation. Therefore, the tree representation allows for a natural way of data integration, avoiding misspecification from restrictive assumptions and overfitting for general models. This paper was accepted by David Simchi-Levi, operations management. Funding: M. Xue’s research is partly support by the National Natural Science Foundation of China [72201257, 72571069, 72371240]. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2023.03967 .

I examine why the prices that consumers pay influence their ratings of products. In the context of hotel reviews on an online booking platform, I find that prices have larger impact on ratings when consumers face larger uncertainty about hotel quality, indicating that prices affect satisfaction through consumers’ quality expectation. On the other hand, I find that the mere fact that the reviewer received higher surplus through a lower price by itself is not a primary reason for a higher rating. I also present suggestive evidence of consumer reciprocation, in which consumers experiencing a high-quality staff service likely become more forgiving of higher prices. I show that introductory pricing can have a substantial impact on the distribution of review rankings among new entrants because of the role of prices as quality signals. Understanding the price effect on ratings is crucial to designing review systems robust to potential distortion. This paper was accepted by Raphael Thomadsen, marketing. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2024.05043 .

This paper develops an economic setting to rationalize the use of the “least absolute shrinkage and selection operator” (LASSO) in estimating asset returns. In this setting, multiple traders engage in trading based on information extracted from historical asset prices. Facing model uncertainty in forecasting asset returns, these traders adopt robust-trading strategies. Within this context, the use of LASSO for estimating asset returns emerges endogenously as an equilibrium outcome. We further extend our analysis to rationalize the application of elastic-net estimation. Although LASSO-type strategies enhance traders’ profits by mitigating competition among them, they also introduce biases in trading decisions, which can adversely affect profitability. This dual effect highlights the nuanced tradeoffs associated with employing such estimation techniques in financial markets. This paper has been This paper was accepted by Will Cong for the Special Issue on AI for Finance and Business Decisions. Funding: This work was supported by the Social Sciences and Humanities Research Council of Canada [Grant 435-2021-0040] and the Bank of Canada [Fellowship]. Supplemental Material: The data files are available at https://doi.org/10.1287/mnsc.2024.06127 .

Many jurisdictions have launched antitrust enforcement and brought in regulation of large tech platforms. The swift and strict implementation of China’s Anti-Monopoly Guidelines for the Platform Economy (Platform Guidelines) provides a quasi-natural experiment to evaluate the impact of antitrust regulation on platform competition. We adopt a difference-in-differences approach to empirically explore the impact of China’s Platform Guidelines on the number of investments and the entry of startups in platform markets. The results show that the Platform Guidelines did not increase entrepreneurship in these affected markets. Rather, entrepreneurship weakened in these markets, with less venture capital investment flowing into them and fewer startups entering these markets. Our study suggests that governments should consider more carefully the potential unintended consequences of antitrust platform regulation. This paper was accepted by D. J. Wu, information systems. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2024.04397 .

In recent years, earnings call slide decks (hereafter, earnings decks) have become increasingly common in the earnings announcements of public firms. In this paper, I examine the characteristics of earnings decks and the determinants of firms’ use of them. More importantly, I study their capital market roles. While I fail to find evidence that earnings decks on average offer information incremental to traditional disclosures, such as press releases and conference calls, they significantly reduce users’ information processing time. My analyses reveal three mechanisms that contribute to this acceleration: inclusion of visuals (e.g., charts and tables), summarization of key information, and release prior to earnings calls. Overall, this study documents a novel aspect of earnings announcements and highlights the role of multimedia elements in reducing information processing costs of corporate disclosures. This paper was accepted by Suraj Srinivasan, accounting. Funding: This work was supported by National Natural Science Foundation of China [Grants 72402110, 72442014]. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2024.04956 .

Biomedical research may uncover insights regarding the interaction of the treatments of a disease with patient covariates. We show how to use such insights to improve the efficiency of adaptive clinical trials for precision medicine by extending ideas from optimal Bayesian learning. We present a model for response-adaptive multi-arm clinical trials that leverages knowledge about the predictive-prognostic covariate structure to accelerate the learning of personalized treatment strategies that obtain the best expected outcomes for post-trial patients. Our base model is a contextual linear bandit for best-arm identification, and outcomes may be observed with delay. We characterize the optimal policy for sequentially allocating treatments to in-trial patients and, because it is hard to compute, propose several computable heuristics based on Bayesian one-step look-ahead techniques. We prove that several of our proposed heuristics are asymptotically optimal in learning treatment strategies. Numerical results based on two case studies motivated by sepsis management show that our heuristics can significantly improve clinical trial efficiency to learn a treatment strategy for precision medicine. We provide extensions that allow for rewards from outcomes of in-trial patients (resolving the exploration-exploitation tradeoff) and for inferring covariate structure using Lasso when biomedical insights on covariate structure are lacking. Our proposed trial design is of interest to funders, designers, and managers of clinical trials. It may also apply to other contextual bandit problems in settings where insights about covariate-treatment interactions are available. This paper was accepted by Stefan Scholtes, healthcare management. Funding: This work was supported by the Marie Skłodowska-Curie Actions under the European Union’s Horizon 2020 research and innovation programme [Grant 676129]. S. E. Chick gratefully acknowledges the support of Dr Simba Gill and Sabi Dau to the INSEAD Healthcare Management Initiative. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2022.02048 .

Health sensing for chronic disease management creates immense benefits for social welfare. Existing health sensing studies primarily focus on the prediction of physical chronic diseases. Depression, a widespread complication of chronic diseases is, however, understudied. We draw on the medical literature to support depression detection using motion sensor data. To connect humans in this decision making, safeguard trust, and ensure algorithm transparency, we develop an interpretable deep learning model: temporal prototype network (TempPNet). TempPNet is built on the emergent prototype learning models. To accommodate the temporal characteristic of sensor data and the progressive property of depression, TempPNet differs from existing prototype learning models in its capability of capturing temporal progressions of prototypes. Extensive empirical analyses using real-world motion sensor data show that TempPNet outperforms state-of-the-art benchmarks in depression detection. Moreover, TempPNet interprets its decision by visualizing the temporal progression of depression and its corresponding symptoms detected from sensor data. We further employ a user study and a medical expert panel to demonstrate its superiority over the benchmarks in interpretability. This study offers an algorithmic solution for impactful social good—collaborative care of chronic diseases and depression in health sensing. Methodologically, it contributes to extant literature with a novel interpretable deep learning model for depression detection from sensor data. Patients, doctors, and caregivers can deploy our model on mobile devices to monitor patients’ depression risks in real time. Our model’s interpretability also allows human experts to participate in the decision making by reviewing the interpretation and making informed interventions. This paper was accepted by D. J. Wu, information systems. Funding: J. Xie and X. Fang are supported by the University of Delaware Research Foundation Strategic Initiatives Grant and Alfred Lerner College of Business and Economics Research Grant, X. Zhao acknowledges financial support from the National Natural Science Foundation of China [Grant 72401172] and the Fundamental Research Funds for the Central Universities [Grant 2023110139, 2023110318]. J. Xie and X. Fang did not receive any form of support from, nor do they have any affiliation with, X. Zhao’s funding sources. Supplemental Material: The online appendices and data files are available at https://doi.org/10.1287/mnsc.2023.04183 .