Biases in demand analysis due to variation in retail distribution

Aggregate demand models typically assume that consumers choose between all available products. Since consumers may be unwilling to search across every store in a given market for a particular item, this assumption is problematic when product assortments vary across stores. Using supermarket scanner data for five product categories we demonstrate that approximately one third of products have limited retail distribution, which account for one fourth of dollar sales. Monte Carlo analysis demonstrates that the level of limited product availability observed in the data can significantly bias the results of aggregate demand models that incorrectly assume all consumers in a given market face the same choice set.     

A diagrammatical template for business market demand estimation

This article describes a template that can help guide managers through a minefield of complex product interactions and demand drivers on the way to producing a realistic demand estimation. It is based on many years of consulting and conducting postmortems to find out what went wrong when demand forecasts have gone awry.A menu of troublesome variables that make up a rogue's gallery of the worst offending causes of estimation error are presented. The premise being, if we are aware of the potential trouble spots in demand estimation, we are likely to avoid some of the more serious problems that plague market projections.A diagrammatical template is presented that shows these troublesome variables and where they enter into the demand evaluation process. Next, the template components are introduced to show how the various product interactions and drivers that affect demand estimation are used and how they impact market projections. Throughout, a list of dos and don'ts are provided along with a sufficient number of real-life disaster stories to reinforce the issues.     

The choice at the checkout: Quantifying demand across payment instruments

Dramatic changes have occurred in the U.S. payment system over the past two decades, most notably an explosion in electronic card-based payments. This shift has led to a series of policy debates driven in part by consumers' choice of payment instruments. Using a new nationally representative survey, we transform consumer responses to open-ended questions into product rankings and estimate a characteristics-based rank-order logit model in order to quantify consumer substitution among payment methods. Our estimates are then used to conduct supply-driven and demand-driven counterfactual experiments in order to estimate market share and cost effects. From a counterfactual experiment in which merchants stop accepting credit cards, we predict merchant costs to decline substantially. Because merchants accept credit cards nonetheless, we regard our finding as evidence either that the credit card networks hold market power, or that merchants experience unmeasured intangible benefits from credit card acceptance. We also predict that contactless debit will take market share from cash, checks, and credit, and that the age/cohort effect alone is unlikely to cause debit card use to increase substantially over a 10-year period.     

Service-based capacity strategy for manufacturing service duopoly of differentiated prices and lognormal random demand

Globalization of production has ushered in vibrant manufacturing service industries whose business is to serve the production needs of product engineering firms. In these engineering–manufacturing chains, risk of capacity supply is an important concern for product firms as manufacturers are normally conservative in capacity expansion, especially when demand uncertainty and investment risk are high. To provide quality manufacturing services, this risk must be taken into consideration in formulating a capacity strategy. This paper presents a competition analysis method based on supply risk for a manufacturing duopoly of differentiated prices and lognormal random demand. A novel service-based demand rationing rule is first proposed. Reaction curves and equilibrium of capacity strategy are next derived. Finally, competition behavior of the duopoly is analyzed.     

Recovering markups from production data

In this paper, I discuss, what I call, the Production-Approach to recovering markups. In contrast to the most popular approach in empirical IO, which relies on demand estimation, this approach requires standard production data while allowing for various price-setting models and puts no restrictions on underlying consumer demand. Using production data together with standard cost minimization allows a researcher to obtain markups in a flexible way. After presenting a brief and selective overview of the literature I contrast the production approach to that of the more popular demand estimation approach. This discussion makes it clear that both approaches face important trade-offs and at a minimum empirical economist should have both techniques as part of their toolbox. The hope is that the use of both methods will only depend on the data at hand and the relevant institutional knowledge, paired with the actual research question we are trying to answer.     

Resale and bundling in auctions

The possibility of resale increases bidders' incentives to jointly reduce demand in multi-object auctions, because resale increases low-value bidders' willingness to pay and reduces high-value bidders' willingness to pay. Therefore, resale may reduce the seller's revenue. In a simple model with complete information, however, allowing resale and bundling the objects on sale are “complement strategies” for the seller (under reasonable conditions)—by bundling and allowing resale the seller earns a higher revenue than by selling the objects separately and/or forbidding resale. We also show why allowing resale may reduce efficiency.     

Recent progress in the empirical analysis of multi-unit auctions

I discuss some recent progress in the empirical analysis of multi-unit auctions, through which a number of important commodities (Treasury bills and bonds, electricity, emission permits, monetary infusions by some central banks) are allocated.     

Distribution contracts to support optimal inventory holdings under demand uncertainty

When suppliers produce products for which demand is uncertain, they face a problem of inducing downstream distributors to stock inventory levels that the suppliers prefer. This paper considers a wide array of alternative supply contracts, each of which consists of a mixture of constant per-unit wholesale prices, buy-back arrangements, and post sale payments contingent on sales made, such as revenue sharing or buybacks. We show that linear supply contracts specifying any combination of two of these three instruments can implement the vertical integrated outcome for a monopoly, thereby generating the supplier's preferred inventory configuration and price distribution. We extend our results to differentiated product oligopoly, demonstrating that each supplier obtains its preferred inventory configuration and price distribution, given the choices of its rival. Distributors choose optimal inventories from the suppliers' standpoint, even if suppliers do not know the distribution of demand uncertainty, and, given the perfect competition among distributors, all profits in the supply chain are captured by suppliers. Thus, suppliers are able to deal with demand uncertainty with remarkably little information about demand, and without the need to control dealer actions in detail. In particular, suppliers need not specify either dealer inventories or resale prices, but instead encourage distributors to order based on information in their possession and to set prices that generate desirable resale price dispersion.     

Using merger simulation models: Testing the underlying assumptions

Merger simulation is now widely used by economists to evaluate the likely competitive effects of a proposed merger. However, the reliability of a given merger simulation depends crucially on the reliability of the data used and the assumptions made. We discuss tests that can be used to assess the reliability of a merger simulation and show how these tests were applied in the context of the Volvo–Scania merger.     

Quantifying price-driven wireless substitution in telephony

For the better part of a decade, a non-trivial and steadily increasing share of households in the United States has come to rely exclusively on wireless technology for their voice communications needs. Aggregate data show clearly (1) that the share of wireless-only households has risen steadily in recent years; while (2) the price of wireless service has fallen substantially relative to traditional landline service. The aggregate data are therefore consistent with the hypothesis that wireless/wireline cross-price elasticities are positive and economically significant. However, econometric corroboration of this conjecture has proven elusive in the existing empirical literature, which has relied on datasets compiled at the turn of the millennium, when wireless substitution was very limited. Partly in response to this dearth of econometric evidence, regulators and competition authorities in the US have generally been reluctant to conclude that wireless voice service represents a meaningful economic substitute for traditional wireline telephony. In the absence of reliable econometric estimates, even the sign of the relevant cross-price elasticities is an open question: The majority of US households maintain both a landline and at least one wireless connection, so it is unclear, ex ante, whether the two services are substitutes or complements. Thus, it is critical to identify consumer behavior at the margin. Using state-level panel data from a relatively recent time period (2001-2007), this study develops and estimates a demand system that permits evaluation of the own-price, cross-price, and income elasticities of demand for wireless and wireline telephony in the United States. A one percent decrease in the price of wireless service is estimated to decrease the demand for fixed-line service by approximately 1.2-1.3%, and the parameter estimates imply that the Slutsky symmetry holds for the demand system. These results substantially exceed prior econometric estimates from the existing empirical literature, and provide evidence that wireless voice service has evolved into a strong economic substitute for traditional landline service. The parameter estimates from the demand system suggest that roughly two thirds of observed landline attrition in the United States over the sample period is attributable to the observed decline in the relative price of wireless service.     

Opportunity cost of spectrum for mobile communications: Evaluation of spectrum prices in Bangladesh

The purpose of this study is to focus attention on the choice of the reserve price made by the telecom regulator to explain the under-performance of spectrum auctions in Bangladesh. The overestimation of the reserve prices for several auctions in Bangladesh including the auction of the 3G – 2100 MHz band in 2013, the 2G - 1800 MHz band in 2015, and the 4G – 1800 and 2100 MHz bands in 2018 is established by comparing the reserve prices to the economic value of spectrum, determined by using the engineering valuation and production function approaches. Possible reasons for the inflation of the reserve price are discussed.     

Managing demand uncertainty through order overplanning

Most techniques for managing demand uncertainty require a certain degree of stability in the environment, since they are completely or partially based on the observation of historical data. When applied to a context characterized by irregular and sporadic demand these techniques show poor performances. In fact, in such a case uncertainty management calls for the gathering of information that directly anticipates future requirements. Although contexts with irregular and sporadic demand have received only minor attention in the past, they are currently gaining ever more importance and extending their occurrence. This paper illustrates and discusses a method, called order overplanning, specifically designed to cope with uncertainty in these environments. It consists of an articulate and coherent set of forecasting procedures, planning principles and slack control techniques. From a Master Production Scheduling (MPS) perspective, order overplanning is similar to hedging and option overplanning: gross requirements are larger than expected demand. The major difference is that order overplanning uses two distinct units in the MPS and forecasting procedures: while the MPS unit is an end item or a module, the forecasting unit is a customer order. This makes order overplanning able to exploit early information generated by each customer during its purchasing process, information that otherwise would be lost. This marked advantage comes to the detriment of an increased effort of integration between Sales and Manufacturing, especially for controlling the slack created to handle uncertainty.The paper first infers the principles and procedures of order overplanning by analysing the case study of an Italian telecommunications manufacturer. Then, it discusses the main advantages and disadvantages of this method, in order to identify the main factors affecting its performances and to determine the planning environments where it fits coherently.     

Risk averse retail pricing with robust demand forecasting

Good demand estimates are the key to effective pricing decision-making. However, they are subject to a high degree of uncertainty due to various factors that are unpredictable or difficult to model, thus making pricing decisions risky. This research provides a simple proposal for a robust optimization methodology that incorporates both demand uncertainty and the decision maker's degree of risk aversion. Uncertainty is explicitly considered for two coefficients of a linear demand function, price expressions are derived, and a criterion is proposed for defining the degree of risk aversion. The resulting model is also applied to an exponential demand case to better reflect a more realistic retail setting.     

The impact of shopbot use on prices and price dispersion: Evidence from online book retailing

Internet price search tools, notably shopbots, have reduced consumers' search costs for price and product characteristics. While a variety of analytic models predict that increased consumer search will lower price levels among competing retailers, there is no consensus in the literature as to how price dispersion will change with increased consumer search. Moreover, there are no papers that have empirically tested these predictions using direct observation of variation in shopbot use over time.This paper examines the impact of changes in shopbot use over time on pricing behavior in the Internet book market. We do this by combining price and clickstream data collected from August 1999 to July 2001 — a period of rapid expansion in shopbot use. We find that a 1% increase in shopbot use is correlated with a $0.41 decrease in price levels and a 1.1% decrease in price dispersion.     

Estimating obsolescence risk from demand data to enhance inventory control—A case study

In this paper obsolescence of service parts is analyzed in a practical environment. Based on the analysis, we propose a method that can be used to estimate the risk of obsolescence of service parts, which is subsequently used to enhance inventory control for those parts. The method distinguishes groups of service parts. For these groups, the risk of obsolescence is estimated using the behavior of similar groups of service parts in the past. The method uses demand data as main information source, and can therefore be applied without the use of an expert's opinion. We will give numerical values for the risk of obsolescence obtained with the method, and the effects of these values on inventory control will be examined.     

Impact of demand uncertainty on supply chain cooperation of single-period products

Cooperation is an approach of improving competitive advantages of a supply chain. A two-echelon supply chain consisting of a manufacturer and a retailer for a single-period product is studied, and retail-market demand uncertainty is described by coefficient of variation. We develop a cooperation mechanism to address the cooperation and its implementation between the manufacturer and the retailer, two market situations are considered: (i) the wholesale price and the order quantity are decision variables, (ii) the wholesale and the retail prices as well as the order quantity are decision variables. In both market situations, our research shows that: (1) the cooperation mechanism can improve the overall channel profits and the supply chain members’ allocated profits, (2) the described cooperation is conditional on retail-market demand uncertainty: it can be implemented if, and only if, the fluctuation of retail-market demand is relatively small and coefficient of variation of retail-market demand does not exceed an upper bound. Impacts of retail-market demand uncertainty on wholesale price, order quantity and/or retail price have also been investigated through analytical and numerical analyses. Although our research is based on the assumption that the manufacturer dominates the supply chain in the non-cooperative situation, which is not the case for most retailer-driven supply chains, this research is still significant on providing guidelines for practitioners in current China mid-level car market that is similar to situations described in the paper.     

Due diligence and the demand for electricity: A cautionary tale

In 1983, the Washington Public Power Supply System (WPPSS) defaulted on tax-free revenue bonds issued to construct two of its five nuclear power plants. In subsequent litigation, the plaintiffs alleged that the bonds had been fraudulently issued because WPPSS and the other defendants should reasonably have known and disclosed that there was substantial uncertainty about the ability to meet the financial obligations created by the bonds. A model using plausible values for demand elasticities, recognized probabilities of events that would increase costs and delay construction and data used by WPPSS to construct demand forecasts for prospective bondholders suggests that such uncertainty was present at the time the bonds were issued. The values for demand elasticities and other parameters of the model were selected after a thorough review of the econometric literature on demand for electricity, beginning with the work of Fisher and Kaysen in 1962.     

Coordinated capacity migration and expansion planning for semiconductor manufacturing under demand uncertainties

Semiconductor industry is very capital intensive in which capacity utilization significantly affect the capital effectiveness and profitability of semiconductor manufacturing companies. Due to constant technology advance driven by Moore's Law in semiconductor industry, multiple production technologies generally co-exist in a wafer fabrication facility with utilization of a pool of common tools for multiple technologies and critical tools dedicated for a specific technology. Because part of the equipment is common for products of different technologies, production managers have limited flexibility to dynamically allocate the capacity among the technologies via capacity migration. The possibility of capacity migration and interrelationship among different technologies make capacity planning difficult under demand and product-mix uncertainties.This paper aims to develop a dynamic optimization method that captures the unique characteristics of rolling demand forecast mechanism to solve capacity expansion and migration planning problems in semiconductor industry. In semiconductor industry, demand forecasts are rolling and updated when the latest market and demand information is available. This demand forecast mechanism makes forecast errors in different time periods correlated. We estimate the validity and robustness of the proposed dynamic optimization method in an empirical study in a semiconductor manufacturing company in Taiwan. The results showed practical viability of this approach and the findings can provide useful guidelines for capacity planning process under rolling forecast mechanism.