Browsing by Author "Swinney, Robert"

This dissertation presents three essays on sustainable and resilient supply chains. The first chapter, joint work with Robert Swinney, examines the profit and environmental implications of supply chain responsiveness. We use a model wherein responsiveness increases fixed and marginal costs, decreases leadtimes, and changes the per-unit environmental impact of production and distribution. We observe that a win-win outcome for the firm and the environment involving responsiveness is most likely to occur when demand variability is high and unsatisfied customers substitute with a less sustainable product.

The second chapter, joint work with Robert Swinney and Nitin Bakshi, studies the value of two vertical control strategies, vertical integration and direct sourcing, in building supply chain resilience. We analyze a model of sourcing of a buyer in a three-tier supply chain where Tier 2 suppliers are disruption-prone. With vertical integration, the buyer purchases one or more Tier 1 suppliers, and with direct sourcing, the buyer purchases raw materials directly from Tier 2 and sells to Tier 1. We show that vertical control in general is most valuable to the buyer when correlation in Tier 2 disruption risk is high, the probability of a disruption is moderate, and disruptions are of moderate severity.

The third chapter, joint work with Robert Swinney, investigates the profit and environmental implications of nearshoring in a decentralized setting. We study a two-tier supply chain with a buyer and a supplier, where the firms first decide whether to reshore or not. Firms' reshoring decisions change their costs, the per-unit environmental impact of their production and distribution processes, and when the buyer has to make procurement or production decisions. We find that if the buyer is more powerful, either a fully offshore or a fully nearshore supply chain has the potential to simultaneously be the equilibrium supply chain configuration and optimal for the environment. If the supplier is more powerful, we observe that either a fully offshore or a hybrid supply chain has this potential.

Consumers have become increasingly sophisticated along multiple dimensions. For example, they may consider whether and when to purchase from a firm based on current and future prices and product availability. Notably, empirical evidence suggests that a significant fraction of consumers do not visit firms that experience frequent stock outs and may postpone their purchase with the hope of paying a lower price in the clearance period. Additionally, consumers also consider product design in their valuation for the product, and are only willing to pay premium prices for products that match design trends and retain value during the product’s lifecycle. In such an environment, retailers should include elements of consumer behavior in their pricing, inventory, channel (e.g. online and offline), and fashion design decisions. In three essays, we address each of these managerial decisions with the aim of helping retailers to understand and control different elements of consumer behavior while providing a better shopping experience for their consumers. In the first essay, we focus on pricing strategies of a firm that implements multiperiod pricing and answer whether consumers necessarily benefit from strategic behavior (timing their purchase) when the firm optimally responds to this type of behavior. We find that many consumers have a lower surplus if they are strategic than if they are myopic. We then develop a model in which consumers choose to become strategic by exerting costly effort, and find that it is possible to increase firm profit, consumer, and social welfare simultaneously by increasing the cost of strategic behavior. In the second essay, considering a product with a decreasing price path, we study how inventory and channel integration impacts whether and when consumers visit the firm by affecting product availability. We show that integration generates value by increasing regular period availability and encouraging more consumers to visit the firm, but it may also harm the firm by encouraging more consumers to delay their purchases. In the third essay, we study joint design, pricing, and inventory decisions of firms facing consumers that have a low valuation for products that are out of style and rapidly lose value during the selling season. We illustrate that firms that implement dynamic pricing should carefully consider the interdependence between their inventory and design decisions. In sum, our findings in this dissertation shed light on the importance of considering implications of consumer behavior in retail operations and provide solutions that benefit firms and consumers.

The last decade has witnessed the rapid growth of many innovative and disruptive business models. This dissertation identifies a few of the unique inefficiencies and challenges that these new business models bring for small entrepreneurs and recommends ways to solve them.

The second and third chapters of this dissertation focus on rewards-based crowdfunding platforms such as Kickstarter and Indiegogo, where entrepreneurs designing new and innovative products solicit monetary pledges from a population of potential contributors. If total pledges exceed a pre-specified funding target, the entrepreneur distributes non-monetary rewards—typically, units of the new product—to these contributors. Otherwise, the contributors are refunded their pledges.

The second chapter explores how an entrepreneur might overcome one of the most significant challenges in crowdfunding: credibly signaling information about the quality of the new product to a pool of small, uninformed contributors. We find that the entrepreneur should signal high quality by setting a high target that is distorted above the full information optimal level. While a separating equilibrium always exists, a pooling equilibrium can only occur under very specific circumstances. We show that the high target affects the quality choice of entrepreneurs and may deter unique, high quality projects. In addition, we discuss how the entrepreneur should modify the signaling strategy when a high target potentially deters backers from pledging due to the cost of participating in a failed campaign.

The third chapter explores how to design such crowdfunding campaigns when contributors choose not just whether to contribute, but also when to contribute. We show that strategic contribution behavior—when contributors intentionally delay their pledges until campaign success is likely—can arise from the combination of non-refundable (potentially very small) hassle costs and the risk of campaign failure, and can explain pledging patterns commonly observed in crowdfunding. Furthermore, such delays hurt the entrepreneur if contributors are distracted, i.e., if they may fail to return to the campaign after an intentional delay. To mitigate this, an entrepreneur can use a simple menu of two rewards with a fixed number of units sold at a low price, and an unlimited number sold at a higher price; this segments contributors over time based on the information they observe upon arrival. Despite its simplicity, such a menu performs well compared to a theoretically optimal menu consisting of an infinite number of different rewards and price levels under many conditions.

The fourth chapter focuses on another business model: subscription box services that deliver shipments of products to consumers at regular intervals for a fixed, per-box fee. Two challenges for providers of such services are acquiring new subscribers and retaining existing ones. We show that providers can respond to these challenges by managing the content of their subscription boxes over time when selling to customers that are heterogeneous along two dimensions: their utility for the contents of each box, and their utility for the service of content curation and delivery. The provider faces a budget for the total value (i.e., the quality) of box contents over a finite time horizon, and must allocate that budget over time to maximize total demand. Allocating more budget to a particular period increases consumer utility for the box—and hence the subscription rate—in that period, but at the expense of reducing the remaining budget for other periods. We show that it is generally not optimal for the service provider to allocate her budget equally and maintain consistent quality over time. If consumer heterogeneity is low, the optimal content strategy is to increase quality over time, which prioritizes retention over initial acquisition (i.e., quality starts low, acquiring few consumers, but increases to induce consumers to continue subscribing). On the other hand, if heterogeneity is high, the optimal strategy is to decrease quality over time, prioritizing acquisition over retention (i.e., quality starts high, acquiring many consumers, but decreases over time, retaining only consumers who highly value the service).

Information goods—such as books, music, and video—have long been sold via a traditional retail model involving physical media (e.g., physical books, CDs, and DVDs) sold "a la carte'' via third party retailers. In recent years, however, innovations in digital distribution technology have enabled new ways of selling, distributing, and consuming information goods. For example, books can now be sold as e-books, and may be consumed via traditional a la carte methods (e.g., when a consumer buys a permanent license to an e-book) or via subscription services (e.g., when a consumer pays a monthly fee for unlimited access to a library of content without ever owning a permanent copy of any books in that library). The rapid growth and change of digital distribution technology has, however, introduced a number of challenges to the management of supply chains for information goods. We examine three problems in this area. First, we analyze a supply chain where a manufacturer produces both physical and digital goods and has a choice between selling through a single retailer who sells both formats or two format-specific retailers. We find that when the manufacturer sells through the single retailer, the supply chain achieves the centralized system outcome by selling zero physical goods and the centralized optimal quantity of the digital good. However, despite this, a manufacturer would prefer to sell through two format-specific retailers rather than through a single retailer, with a strictly positive quantity of the operationally inferior physical good, to the detriment of total supply chain efficiency. Additionally we find that consumers and society are both better off when the manufacturer sells through independent retailers. Next, we analyze selling decisions for supply chain of a digital content platform and the two creators (low and high quality) who sell their content through the platform. The creators have a choice of selling their content a la carte and/or via subscription and being paid via revenue sharing contracts. We find that the platform cannot induce only the high quality consumer to sell via subscription which means that the subscription offerings are always weakly lower in quality than a la carte offerings. We also find that in many cases, to maximize profit, the platform should either induce only the low quality creator to sell via subscription, or it should shut down a la carte sales altogether; inducing high quality on the subscription service is excessively costly. This effect can be mitigated—and inducing high quality on the subscription service can be optimal for the platform—in the presence of a large “subscription only” consumer segment. In the third paper, we explore the optimal design of revenue splitting rules that feasible (i.e. induces all creators to join the subscription service), fair (i.e. allocates revenue only based on amount of consumption generated by each creator) and optimal (i.e. both fair and feasible at the lowest cost possible to the platform) and allows a platform to maximize its own revenue while inducing a wide variety of high quality content on the service. We show that a splitting rule that is quadratic in the consumption of each creator's product is optimal while a linear rule is not and that a linear rule can actually perform arbitrarily bad under some circumstances.