Information Technology (IT) and IT-enabled applications are changing consumer behavior as well as how firms interact with their customers. I am interested in understanding the impact of these changes on consumers and firms. My research investigates questions that fall broadly into four application areas with significant overlap: (1) Information and Communication Technology for Development (ICT4D), (2) IT-enabled Business Models, (3) Digital Financial Services, and (4) Supply Chain Coordination. In each of these areas I apply empirical methods to rigorously identify areas in which IT is beneficial, as well as to make policy suggestions to mitigate any detrimental effects. My goal in each of these analyses is to provide a causal link wherever possible.
Information and Communication Technology for Development (ICT4D) is a research area that examines how the introduction of mobile phones and other information and communication technologies (ICTs) have improved the lives of individuals living in developing economies. I focus on how IT-enabled applications reduce market frictions and information asymmetries resulting in net welfare gains, especially for individuals who are the most marginalized. My specific goals in this stream are 1) to rigorously measure the causal impact of different initiatives, and 2) to determine how the successful initiatives can be quickly improved and scaled to achieve maximum benefit in a short amount of time with minimal resources while minimizing any unintended negative consequences. My research informs managers and policy makers which strategies lead to more efficient markets and supply chains, ultimately resulting in an overall reduction in the poverty rate.
Access to ICTs is widely known to improve market efficiency. However, before Parker, Ramdas, and Savva (2016) it was unclear whether access to timely and accurate information provided through IT-enabled applications has any additional impact. Using a detailed dataset from Reuters Market Light (RML), a text message service in India that provides daily price information to market participants, we find that this information reduces the geographic price dispersion of crops in rural communities by an average of 12%, over and above access to mobile phone technology and other means of communication. Policy makers and international aid organizations should take these results into consideration when deciding how to allocate funds aimed at improving welfare in developing countries. Complementing ICT infrastructure projects with subsidies for farmers that subscribe to price information services will lead to further reductions in price dispersion and the associated improvements in social welfare. Parker and Weber (2011) provides a fuller description of the technology RML uses to collect and disseminate the price information. We highlight the difficulties involved with setting up a successful and impactful information system where low-tech solutions are the primary price discovery method, and discuss the way in which RML overcame these challenges.
Acimovic et al. (2018) takes a more hands-on approach to ICT4D research by helping a company optimize their service. Specifically, we explore how a text message-based decision support system can be used to improve agent physical/electronic cash ratios at the beginning of each day in order to reduce electronic cash stockouts. To do so, we ran a field experiment altering the type of guidance provided in daily text messages and the type of training provided. We find that agents who are trained in person and receive an explicit recommendation as to how much electronic currency to have at the beginning of a day reduce stockouts by 2.8-3.8 percentage points (on a baseline of about 31.6%). We also find significant agent-level heterogeneity in the extent to which agents benefit from the treatment with agents that experience substantially more customer cash deposits (as opposed to cash withdrawals) improving the most.
IT-Enabled Business Models allow businesses to alter both the information presented to a consumer and the information that companies can gather about consumers. In Mejia and Parker (2018), we explore how this information can result in biased behavior in ridesharing platforms. After early work demonstrated that bias exists, platforms responded by removing information about the rider’s gender and race from the ride request presented to drivers. However, following this change, bias may still manifest after a request is accepted, at which point the rider’s name and picture are displayed, through driver cancelation. 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 on ridesharing platforms. We investigate this through a large field experiment using a major ridesharing platform in North America. Our results confirm that bias at the ride request stage has been eliminated. However, at the cancelation stage, racial and LGBT biases are persistent, while biases related to gender appear to have been eliminated. We also explore whether dynamic pricing moderates (through increased pay to drivers) or exacerbates (by signaling that there are many riders, allowing drivers to be more selective) these biases. We find a moderating effect of peak pricing, with consistently lower biased behavior. Our results highlight the importance of carefully considering not just what information to provide, but also when that information should be communicated.
Other work in this area explores how the sunk cost fallacy impacts bidding behavior in pay-to-bid auctions. Jindal et al. (Work in Progress) explores the role the sunk cost fallacy plays in keeping an innovative retail website open. The website conducts penny auctions where individuals pay a non-recoverable bidding fee to increment the current price by a penny. The implementation we study is an interesting place to explore the sunk cost fallacy because some bids are “free”–the bidder did not explicitly pay for them but either won them in a previous auction or was given them by the company. The introduction of free bids means that we can separate the financial and psychological mechanisms behind the sunk cost fallacy.
Digital Financial Services are the result of integrating IT into mainstream financial services around the globe. My research in this area explores how the new systems have changed the way individuals and companies consume financial services, and evaluates new IT-enabled initiatives that may fundamentally change financial services.
Parker and Weber (2014) empirically tests how characteristics of brokers can be used to predict their order routing decisions to two competing electronic exchanges. In contrast to previous research on markets, we find that affiliation is the most important factor driving the percent of orders sent to an exchange and network effects are not significant predictors of brokers’ order routing practices. The results highlight the importance of factors beyond traditional network effects in explaining new market success/failure and the need for exchanges to retain a group of dedicated users.
Recent changes in the financial services industry are being termed as the “Fintech Revolution.” Gomber at al. (2018) presents a new Fintech innovation mapping approach that enables an assessment of the extent to which there are changes and transformations in four key areas of the financial services industry. Through the lens of this mapping, we analyze: (1) operations management in financial services, and the changes that are occurring there; (2) technology innovations that have begun to leverage the execution and stakeholder value associated with payments settlement, cryptocurrencies, blockchain technologies, and cross-border payment services; (3) multiple fintech innovations that have impacted lending and deposit services, peer-to-peer (P2P) lending, and the use of social media; (4) issues with respect to investments, financial markets, trading, risk management, robo-advisory, and related services that are influenced by blockchain and fintech innovations.
Supply Chain Coordination focuses on making the entire supply chain, or at least parts of it, more efficient. Most of my work in this area examines these problems in the for-hire trucking industry, where there is no centralized national or even regional market for services making planning transportation costs a difficult task. Furthermore, contracts in the industry are unusual in that, although they establish prices for different services, there is typically no legally binding obligation or penalty for either party to offer or accept a load. When a load is rejected by all contract carriers, shippers must turn to the hyper-local spot market, which can significantly increase supply chain costs. In Scott, Craighead, and Parker (2018; private working paper), we explore how a small change to the contract—adding a weekly minimum load agreement provision—reduces carriers’ rejection probabilities. Using EDI data from a major national manufacturer detailing every offer for a load and whether it was accepted or rejected, we find that the more explicit contracts reduce rejections. Further, we find that as market conditions change in the supplier’s favor, the effectiveness of more explicit contracts weakens relative to commitment. This paper is important as it demonstrates how a small piece of information added to an unenforceable contract can help to reduce supply chain costs. This work builds on Scott, Parker, and Craighead (2017) where we examined the economic, operational, relationship-based reasons for refusing to provide a contracted service using a detailed transactional data set of a large national shipper. We find key operational and economic factors to be drivers of freight rejection and the shipper-carrier relationship to be a deterrent to freight rejection. We also find that primary and secondary carriers respond differently to these operational and economic factors.
My other work in this area, Moritz, Narayanan, and Parker (2018), explores how cost increases due to behavioral ordering are shared in a supply chain where some of the echelons have automated (computer/algorithmic) ordering systems in place but a human is placing orders in one echelon. We find that costs for the supply chain increase relative to one where ordering decisions in all four echelons are made by computers, creating an argument for more decision support and/or complete ordering control to be transferred to monitored computerized ordering systems.