Providing transparency into operational processes can change consumer and worker behavior. However, it is unclear whether operational transparency is beneficial with potentially biased service providers. We explore this in the context of ridesharing platforms where early evidence documents bias similar to what has been observed in traditional transportation systems. Platforms responded by reducing operational transparency through removing information about riders’ gender and race from the ride request presented to drivers. However, following this change, bias may still manifest through driver cancelation after a request is accepted, at which point the rider’s picture is displayed. Our primary research question is to what extent a rider’s gender, race, and perception of support for lesbian, gay, bisexual, and transgender (LGBT) rights impact cancelation rates. We investigate this through a large field experiment on a major ridesharing platform in Washington, DC. By manipulating rider names and profile pictures, we observe drivers’ behavior patterns in accepting and canceling rides. Our results confirm that bias at the ride request stage has been eliminated. However, after acceptance, racial and LGBT biases are persistent, while we find no evidence of gender biases. We also explore whether peak times moderate (through increased pay to drivers) or exacerbate (by signaling that there are many riders, allowing drivers to be more selective) these biases. We find a moderating effect of peak timing, with lower cancelation rates for non-Caucasian riders. We do not find a similar moderating effect for riders that signal support for the LGBT community. This paper was accepted by Vishal Gaur, operations management.

The opioid crisis has ravaged the United States, taking 69,000 lives in 2020, with prescription opioids accounting for 98% of opioid abuse. Although this epidemic is often considered a White public health crisis nationally, overdose deaths among people of color doubled from 2017 to 2019. As research has shown that the crisis was fueled by excessive supply from the pharmaceutical industry, several individual firms have received significant public criticism. However, we find evidence that the scope of the blame transcends individual actors to indict the very structure of complex supply chains, which may have exacerbated the crisis by dispensing significantly more opioids. In specific, we posit that supply chain complexity allowed mass quantities of opioids to escape detection by the Drug Enforcement Administration (DEA). Further, we find new evidence showing the greater impact of complexity on opioid dispensing in non-White communities, which underscores their exclusion from the public discourse and governmental response surrounding the crisis and suggests possible racial bias in the DEA’s regulatory policies. Our analysis was made possible by the 2019 release of the DEA’s Automation of Reports and Consolidated Orders System database, which logged every shipment in the US opioid supply chain from 2006 to 2014. Using a fixed effects model, we find that a one-unit increase across three dimensions of supply chain complexity is associated with a 16% increase in opioid dispensing. This effect is intensified in non-White communities, where a 10% increase in the non-White population is associated with a 3.39% (1.33%) increase in opioid dispensing by pharmacies that have supply chains with high (average) complexity. To verify that high-complexity pharmacies’ excess dispensing supplied non-medical/recreational demand, we exploit the reformulation of OxyContin (designed to prevent recreational use) as an exogenous shock to the market. In a novel approach, we leverage the fact that different pharmacies received their first shipment of reformulated OxyContin at different times and use a difference-in-differences model to estimate the heterogeneous effect of the shock on dispensing. As the reformulated OxyContin stifled demand, high-complexity pharmacies experienced a 15.31% greater reduction in dispensing compared to lower-complexity pharmacies, suggesting that their excess dispensing was indeed satisfying non-medical/recreational demand.

Problem description: Measuring quality in the service industry remains a challenge. Existing methodologies are often costly and unscalable. Furthermore, understanding how elements of service quality contribute to the performance of service providers continues to be a concern in the service industry. In this paper, we address these challenges in the restaurant sector, a vital component of the service industry. Academic/practical relevance: Our work provides a scalable methodology for measuring the quality of service providers using the vast amount of text in social media. The quality metrics proposed are associated with economic outcomes for restaurants and can help predict future restaurant performance. Methodology: We use text present in online reviews on Yelp.com to identify and extract service dimensions using nonnegative matrix factorization for a large set of restaurants located in a major city in the United States. We subsequently validate these service dimensions as proxies for service quality using external data sources and a series of laboratory experiments. Finally, we use econometrics to test the relationship between these dimensions and restaurant survival as additional validation. Results: We find that our proposed service quality dimensions are scalable, match industry standards, and are correctly identified by subjects in a controlled setting. Furthermore, we show that specific service dimensions are significantly correlated with the survival of merchants, even after controlling for competition and other factors. Managerial implications: This work has implications for the strategic use of text analytics in the context of service operations, where an increasingly large text corpus is available. We discuss the benefits of this work for service providers and platforms, such as Yelp and OpenTable.