This study investigates the impact of mobile hailing technology adoption and network integration on traditional taxi drivers' performance and efficiency. Utilizing Two-Way Fixed Effects (TWFE) and Callaway and Sant’Anna (CS) difference-in-differences models, we analyze data from over 5,000 drivers across six years in Chicago. Our findings reveal that technology adoption significantly enhances performance metrics, including an average 17% increase in income and number of trips, primarily driven by longer working hours. However, efficiency measures do not show improvement and, in some cases, are negatively affected by higher usage rates. Surprisingly, when incorporating the network effect of market integrator, drivers who adopt technology and rent vehicles from partner companies experience even greater performance gains, such as a 25.6% increase in income. Nonetheless, the combination of high usage rates and network effects can lead to reduced operational efficiency, as drivers may become overly reliant on the app and lose focus on traditional street-hailing opportunities. These results highlight the complex interplay between technology adoption and network integration, suggesting that while mobile hailing technology can boost driver performance, its efficiency benefits are contingent upon effective network support. This research contributes to the understanding of how traditional taxi drivers can navigate the competitive landscape shaped by ride-sharing platforms and provides insights for service providers and policymakers aiming to foster sustainable growth and operational efficiency in urban transportation.

Consumers’ willingness-to-pay for a durable good depends upon its perceived longevity, i.e., the duration for which the product continues to function properly. However, because consumers cannot observe longevity prior to purchase, it is difficult for manufacturers to capitalize on investments that enhance product longevity. In this paper, we demonstrate that the leasing market can provide a credible mechanism for signaling product longevity. In our analytical model, a manufacturer both leases and sells a durable good. Leased units are returned to the manufacturer after the lease period ends and then leased again in the secondary market as long as they continue to be functional. When consumers cannot observe longevity, a separating equilibrium arises in which a manufacturer signals high longevity by reducing the lease price and allocating a disproportionate share of production to the leasing market (i.e., uses a high “leasing intensity”). Consumers benefit from this credible signaling because lessees now incur lower usage costs and buyers now avoid overpaying for products with low longevity. In fact, consumers, on average, are better off in the signaling equilibrium than if longevity were directly observable. Our finding that signaling enhances manufacturers’ incentives to invest in product longevity suggests this could be an important mechanism for promoting sustainability initiatives.

This paper examines the effects of service experience ratings (SERs) in service industries. SERs are defined as consumers’ assessments of a service encounter (typically on a 3-, 5-, or 10-point numeric scale) that are made immediately after the encounter. We conceptualize SERs as being a mechanism for retailers to observe how much effort employees expend on the provision of experience quality. Using a principal-agent analytical model in which the worker provides credence quality for altruistic reasons, we investigate whether a retailer should solicit SERs, how SERs influence a worker’s expenditure of effort on experience and credence qualities, and the impact of SERs on customer retention. SERs present a fundamental tradeoff between the benefits from heightened customer satisfaction (which generates future profit) and larger costs (due to monitoring costs and additional compensation to induce the worker to exert more effort). We find that soliciting SERs is profitable for a retailer when monitoring costs are low, the worker’s effort closely matches customers’ preferences for experience attributes, consumers expect a low service quality, and satisfied customers generate high future profit. However, SERs may result in an inefficiently-large reliance on experience quality to generate customer satisfaction. The nuanced effects of SERs lead to several surprising results. For example, it may be optimal for the retailer to respond to higher expectations about quality by providing a lower-quality service. Also, although SERs only assess experience quality, it is possible that SERs may become more beneficial when consumers more highly value credence quality. Another intriguing result is that there are situations in which SERs do not enhance a retailer’s profit even if were costless to collect this information. For example, higher worker altruism may eliminate the incentive to collect SERs and, as a result, lead to fewer customers being satisfied with the service.

As home-sharing platforms mature, hosts face intense competition, necessitating a deeper understanding of how guests navigate the complex array of signals presented online. While prior research has examined individual signals, the interplay between initial, easily visible cues (First-Impression Signals) and deeper, diagnostic information (Ancillary Signals) remains underexplored. This study addresses this gap by developing and empirically testing a two-stage framework that models how guests sequentially process these layered signals to form booking intentions using structural equation modeling and sentiment analysis on a large dataset of 99,034 U.S. Airbnb listings and 3.4 million reviews. Our findings reveal a dynamic process where Ancillary Signals refine initial assessments formed from First-Impression cues and also exert a direct, independent influence on booking intention. We confirm the diminishing relevance of traditional signals (overall ratings, Superhost status) and highlight the rising importance of specific ancillary details, such as location sentiment in reviews, property amenities, and, counterintuitively, stricter cancellation policies. This research contributes a unified framework for understanding signal interplay in P2P platforms, emphasizes the critical dual role of ancillary information, challenges assumptions about conventional quality cues, and offers actionable insights for hosts and platform managers seeking to enhance trust and optimize performance in the sharing economy.

On platforms like Airbnb, visual signals play an important role in shaping expectations, but the ease of image manipulation raises concerns about the credibility of these signals. Leveraging a large panel dataset of over 400,000 properties, more than 18 million guest reviews, and over 1.2 million listing images spanning 2015 to 2025, I investigate the market consequences of this phenomenon. To identify instances of tampering at scale, I employ a state-of-the-art, two-step multi-modal AI framework. First, a Domain Tag Generator classifies the manipulation type (Photoshop or AI); this output then guides a Multi-Modal Large Language Model to conduct a deep authenticity assessment based on both pixel-level and semantic inconsistencies. To measure consumer perception, I analyze review text with a BERT-based model, creating a continuous sentiment score for each review from the expected value of its predicted star-rating probabilities. Then, using matrix completion method, well-suited for high-dimensional panel data with latent confounders, I estimate the causal effect of manipulated images on guest satisfaction and property performance. This research provides the first large-scale systematic evidence on the market outcomes of visual misrepresentation in the experiential markets, where consumption is non-returnable, by quantifying the impact of image manipulation on consumer perceptions and firm outcomes.