6월 4주차

In 2018, the Supreme Court of the United States struck down the Professional and Amateur Sports Protection Act (PASPA), ending a nearly 30-year federal ban on sports betting and paving the way for dozens of states to legalize such operations. The impact has been pronounced, with the sports betting market growing nearly 27% year on year since the decision. Yet, while our understanding of the financial effects of these markets is beginning to emerge, the downstream consequences of the legalization of sports betting remain understudied. In this paper, we examine the impact of legalization of both offline and online sports betting on the well-being of individuals in those jurisdictions. We focus on two outcomes: the number of calls to the National Problem Gambling Hotline and the number of suicides reported per state as a result of legalization of sports betting. Using a difference in differences approach to exploit the phased legalization of sports betting at the state level, results indicate that while the number of calls associated with problem gambling is uncorrelated with the legalization of physical sportsbooks, it is strongly correlated with the legalization of online sports betting. Further, results suggest that the legalization of online betting is correlated with an increase in suicides, an outcome historically associated with problem gambling. Finally, we observe that these deleterious effects manifest more strongly for particular groups (viz. young, unmarried, and relatively less educated men). Our work informs policymakers of the implications of legalization of sports betting, thereby adding to the growing research addressing this important topic.

Air pollution imposes significant cognitive strain and economic costs, yet its effects on digitally mediated, cognitively demanding activities remain underexplored. We examine its impact on digital reading, a context where key digital affordances—ubiquitous access, low switching costs, and personalization—fundamentally distinguish it from print reading. These affordances can buffer immediate disruptions from pollution but also reshape user responses to cognitive fatigue in ways unique to digital environments. Drawing on Cognitive Load Theory, we argue that air pollution depletes cognitive resources and prompts users to adjust their engagement strategies. Using individual-level behavioral data from a major digital reading platform and a classic spatial regression discontinuity design exploiting the Qingling Mountains–Huai River heating policy boundary, we identify four adaptive behaviors facilitated by digital affordances: genre downgrading, familiarity seeking, increased fragmentation, and slower reading pace. Mediation analysis shows that these adaptations fully mediate the effect of pollution-induced cognitive strain on engagement. This suggests that, while they offset the direct effect of cognitive strain, they reduce the utility of reading and thereby lead to lower overall engagement. Through these adaptive behaviors, a one-unit increase in the Air Quality Index reduces reading time by 0.7% and spending by 1.2%. This study integrates perspectives from digital affordances, air pollution, cognitive effects, and reading to uncover affordance-enabled mechanisms through which environmental stressors reshape digital behavior. The results offer design implications for preserving user engagement and policy insights for broadening the assessment of pollution’s societal costs.

The transition of display ad exchanges from second-price to first-price auctions has raised questions about its impact on revenue. Auction theory predicts revenue equivalence between these two auction formats under standard assumptions. However, display ad auctions differ from standard auction models in at least two important ways. First, automated bidding agents cannot easily derive equilibrium strategies in first-price auctions because distributional information about competitors' values, or even the number of competitors, is often unavailable. Second, due to principal-agent problems, bidding agents often optimize return on investment (ROI) rather than quasilinear payoff. The literature on learning agents for real-time bidding is growing because of the practical relevance of this setting. However, whether such learning agents converge to equilibrium is an open question: learning dynamics in games can cycle, become chaotic, or generate off-equilibrium outcomes. Recent experiments suggest that learning agents may also converge to collusive low-price outcomes. Since bidders' underlying values are typically unobserved, it is difficult to determine from field data alone whether observed bids are consistent with equilibrium behavior. In this paper, we derive equilibrium predictions and study the convergence behavior of widely used online learning algorithms in a stationary benchmark model of display-advertising auctions. We also leverage recent developments in equilibrium computation to obtain equilibrium predictions in settings where analytical solutions to the governing differential equations are unavailable. In the stationary benchmark environments we study, the learning algorithms do not exhibit systematic bid suppression and instead move toward the computed equilibrium. The benchmark identifies a theoretically important channel through which auction-format changes can affect revenues when bidders optimize ROI rather than quasilinear payoff: lower first-price revenues can arise from non-collusive ROI-based equilibrium behavior in a canonical auction model. We show that, in equilibrium, second-price auctions achieve higher expected revenue than first-price auctions with ROI-maximizing bidders. These results do not imply that algorithmic collusion is unlikely in real display-advertising markets, but they show that lower first-price revenues are not uniquely diagnostic of collusion and may also reflect bidder objectives.

Public transit is a major setting in which individuals spend time on smartphones, yet relatively little is known about how contextual factors in transit environments are associated with mobile usage. In this study, we examine how physical crowdedness is related to organic (self-initiated) smartphone use and how this relationship varies across usage types, individuals, and contexts. We draw on Personal Space Invasion Theory (PSIT) as a conceptual lens to organize potential behavioral responses to crowded environments, emphasizing a sequence involving arousal, attribution, and compensatory behavior. Using detailed mobile usage data from over 200,000 individuals commuting on a major subway system in Qingdao, China, we document systematic associations between crowdedness and organic mobile usage. In particular, crowdedness is positively associated with overall usage, with stronger relationships for social functionalities, among users with shorter prior trips, during off-peak periods, and when individuals are surrounded by different-age groups. To complement the observational analysis, we conduct an online survey with a randomized design, which provides supportive evidence on how crowded environments are perceived and how such perceptions are associated with usage intentions. By shifting the focus from promotion-induced behavior to organic mobile usage, this study provides new empirical evidence on how everyday environmental conditions are associated with digital engagement, and highlights rich heterogeneity in these relationships across contexts.

Emerging licensing regimes that generate revenue from game sales, along with gamers’ consideration of video game purchases in their platform adoption decisions, call for a reexamination of traditional models of platform competition. We develop a novel model of competition between ecosystems, defined by platforms and their console-specific subsidiary markets for video games, to incorporate post-adoption factors into platform pricing decisions. Counterintuitively, our results show that access to additional surplus from game sales intensifies competition, forcing platforms to further subsidize the gamer side. In the absence of royalties, publishers’ licensing fees remain unaffected even when the subsidiary market is considered; however, these fees increase under royalty transfers, suggesting that money-side considerations dominate in ecosystem competition. When publishers multihome, the role of the subsidiary market becomes more pronounced: gamer-side subsidies deepen while platforms impose higher licensing fees. The subsidiary market also induces greater multihoming than predicted by traditional licensing models. Surprisingly, the effect of royalties on platforms becomes non-monotonic, in contrast to scenarios where all publishers singlehome. This leads to a threshold beyond which platforms may no longer prefer multihoming, diverging from conventional models where multihoming is typically advantageous. Overall, our findings highlight that platforms’ strategic responses to gamers’ post-adoption behavior and royalty-based licensing structures cannot be inferred through simple extensions of existing frameworks.

Online dating and matrimonial matching platforms suffer from a common problem: Men far outnumber women, making it harder for parties to find matches. Men face inordinately low odds of making a connection with women, whereas women face a barrage of messages from unsuitable men. Gender skew causes women to leave the platform, worsening the problem. In a field experiment with a large Indian matrimonial platform, we test a proposed solution called "gender gating": a default setting where we restrict the visibility of women profiles to men who “match” their requirements, based on accepted social norms on age, education, and income. Women can choose to override these defaults. The logic is simple: Limiting visibility cuts unwanted attention at source instead of requiring women to exert effort in screening messages. We see positive results: Among treated women, incoming requests fell 7%, and matching efficacy (incoming requests converting to matches) rose 72%. Women initiated 75% of contacts themselves, indicating growing comfort with the platform. Importantly, outcomes for men did not worsen because they saw profiles that were suited to them per social norms. In all, our approach shows that platform design grounded in local norms is a more precise lever than blanket regulations.

Organizations are increasingly integrating artificial intelligence (AI) into traditionally human-driven evaluation processes, such as recruitment and performance evaluation. Yet the complexity of the underlying AI algorithms often renders these processes opaque, limiting users’ understanding and potentially inducing stress and other behavioral changes. In response to these challenges, transparency is often advocated as a mitigative approach, but its effects remain somewhat ambiguous. While transparency in algorithmic evaluation may mitigate interviewees’ stress in the recruitment context, it may also incentivize opportunistic impression management (IM), thereby engendering a transparency paradox. Focusing on this tension, we theorize that while algorithmic evaluation, relative to human evaluation, may heighten stress and deceptive IM in the recruitment process, transparency has the potential to mitigate these disruptions. We test this across eight studies, including two main experiments and a quasi-field replication. In Experiment I, we found that while algorithmic evaluation increased interviewees’ stress and deceptive IM relative to human evaluation, transparency counteracted these effects, aligning them closely with human evaluation. In Experiment II, we study how these effects caused deviations in downstream interview performance when such performance is assessed in human-only, AI-augmented (wherein human decision-makers are assisted by AI scores), or fully automated decision-making configurations. Although transparency consistently reduced deviations in stress and IM from human evaluation for interviewees, its corrective effect on interview performance was constrained on the evaluative side: when AI scores were present, evaluators anchored on them, discounting their own judgment, even as the AI scores failed to distinguish honest from deceptive behaviors. Together, these findings disentangle the dual-edged effects of transparency in algorithmic evaluation, showing that its benefits depend not only on how users respond, but also on how evaluation is conducted. This work advances research on algorithmic decision-making and socio-technical systems, and offers implications for organizations, policymakers, and AI developers.

Cloud platforms are critical infrastructure for digital economy. This study explains how large cloud providers use prices and service quality to allocate nonstorable computing time among customers with different willingness to pay and delay sensitivity. Pooling many uncorrelated workloads makes utilization predictable, reducing the need for idle backup capacity and improving operating efficiency. The predictability, however, allows providers to profitably throttle lower-tier customers through slower processing or higher interruption risk while keeping the whole market served. For cloud managers, the analysis shows how cross-region pooling, transparent interruption policies, and auction-based spot pricing can improve utilization, segment demand, and guide capacity planning. For customers and policymakers, the results identify why throttling persists: high switching costs and proprietary ecosystems weaken competitive pressure. Policies that improve workload portability, reduce data-egress frictions, and increase data interoperability can benefit the consumers and society without sacrificing the scale economies of large cloud networks. Structural remedies that reduce scale may curb market power but risk undermining operational efficiency.

The rise of new entrants in content-streaming markets has intensified competition for established platforms like Netflix, prompting them to reassess their pricing and content strategies. In this research, we examine an incumbent streaming platform that offers both exclusive and non-exclusive content, exploring its optimal pricing response to the entry of a competitor providing only non-exclusive content under a subscription or usage-based pricing model. We develop an analytical framework to study the incumbent's strategic choices, focusing on key factors such as consumer heterogeneity, consumption intensity, content acquisition costs, and the entrant's pricing strategy. Our findings reveal that the incumbent does not always differentiate its pricing strategy from the entrant and, under specific conditions, matches its pricing approach. Entry is governed by a cost threshold that depends on consumer valuation and the incumbent’s pricing response, and higher valuation does not always strengthen the entrant’s incentive to enter, as incumbents can deter participation through strategic pricing. Moreover, we show that increased consumption can harm the entrant's profitability when relying on usage-based pricing. We also find that entrants may invest more in content than incumbents when acquisition costs and exclusivity appreciation are low, and that incumbents do not always attain a positional advantage, as market conditions such as valuation and content costs significantly affect profitability. Our work offers actionable insights for streaming platforms navigating competitive markets. Platforms must align their pricing and content strategies with consumer behavior, competitive dynamics, and content characteristics. Our findings provide a robust framework for understanding how incumbents and entrants can optimize their strategies in the evolving landscape of streaming markets, contributing to both academic literature and managerial practice.

Ideators increasingly turn to generative artificial intelligence (GenAI) to improve the creativity of their ideas and reduce the cognitive effort required to refine them. This collaboration is based on fine-grained configurations of human-AI agency that allow for partial automation and augmentation of the creative-ideation process. In this study, we examine how representational differences in AI-generated inputs into human ideation in the form of textual and visual concretizations influence the creativity of the jointly produced ideas and the effort ideators must expend. These AI-generated concretizations transform initial raw ideas into more mature representations that ideators can inspect, interpret, and evaluate. In an online experiment, we found that AI-generated textual concretizations improved idea creativity by 18% relative to AI-generated visual concretizations but required 30% greater effort from ideators. This effect was most pronounced for more mature ideas (i.e., specific and actionable). For very immature ideas, visual concretizations enhanced idea creativity without increased effort. We explain these differences through different configurations of human-AI agency in ideation. Textual concretizations correspond to augmentation: GenAI produces coherent textual concretizations that ideators must interpret and complete with their imagination. The material agency of GenAI is matched by human agency, increasing idea creativity but requiring greater effort. As ideas mature, richer textual concretizations provide greater substance for creative elaboration, boosting both idea creativity and ideator effort. In contrast, the use of visual concretizations follows an automation logic. GenAI exerts material agency by autonomously specifying all required details for a visual concretization. While this can boost creativity for very immature ideas by offering stimulating details for creative exploration, these details become constraining as ideas mature. This constrains human agency and ideators’ abilities to integrate novel elements into ideation. As its main contribution, our paper shows that representational differences in AI-generated concretizations shape idea creativity and ideator effort by producing distinct configurations of human-AI agency.