nep-dcm New Economics Papers
on Discrete Choice Models
Issue of 2020‒08‒31
seventeen papers chosen by
Edoardo Marcucci
Università degli studi Roma Tre

  1. Estimating the Social Value of Specific Crop Diversity Conservation Plans: Do Czechs Care More About Conserving Hop, Wine or Fruit Tree Varieties? By Nicholas Tyack; Milan Scasny
  2. Willing to Pay? Spatial Heterogeneity of e-Vehicle Charging Preferences in Germany By Wolff, Stefanie; Madlener, Reinhard
  3. Why do (or don't) people carpool for long distance trips? A discrete choice experiment in France By Guillaume Monchambert
  4. Consumer Preferences for Genetically Modified Organisms in Cape Town: A Choice Experiment Approach By Benjamin Dovey; Herbert Ntuli
  5. Tractable Constrained Optimization over Multiple Attributes under the Multinomial Logit and Nested Logit Model By Hongzhang Shao; Anton J. Kleywegt
  6. Imperfect Mobility By Cai, Zhengyu
  7. The Decision-Conflict and Multicriteria Logit By Georgios Gerasimou
  8. Scalable Bayesian estimation in the multinomial probit model By Ruben Loaiza-Maya; Didier Nibbering
  9. Willingness-to-Pay for Reshuffling Geographical Indications By Monia Saïdi; Jean-Sauveur Ay; Stephan Marette; Christophe Martin
  10. Concordance-based predictive measures in regression models for discrete responses By Denuit, Michel; Mesfoui, Mhamed; Trufin, Julien
  11. Alternative Methods for Studying Consumer Payment Choice By Oz Shy
  12. Service characteristics and the choice between exports and FDI: Evidence from Belgian firms By Leo Sleuwaegen; Peter M. Smith
  13. Partitioned Pricing and Consumer Welfare By Kevin Ducbao Tran
  14. Strategy-proof Choice under Monotonic Additive Preferences By Eric Bahel; Yves Sprumont
  15. Does Voting Solve Intergenerational Sustainability Dilemma? By Shun Katsuki; Yoichi Hizen
  16. The Role of Government in the Market for Electric Vehicles : Evidence from China By Li,Shanjun; Zhu,Xianglei; Ma,Yiding; Zhang,Fan; Zhou,Hui
  17. Two-Stage Majoritarian Choice By Sean Horan; Yves Sprumont

  1. By: Nicholas Tyack (Graduate Institute of International and Development Studies, Geneva, Switzerland); Milan Scasny (Institute of Economic Studies, Faculty of Social Sciences, Charles University, Prague, Czech Republic & Charles University Environment Centre, Prague, Czech Republic)
    Abstract: We use a discrete choice experiment to elicit the preferences of Czech adults ages 18 to 69 (n=805) for the conservation of wine, hop, and fruit tree varieties. In addition, we also elicit the preferences of a smaller sub-sample consisting solely of respondents from South Moravia (n=463), an agricultural region of the country. Estimating a mixed logit model, we find a strong public preference for fruit tree conservation and derive a mean willingness to pay (WTP) for the conservation of fruit tree varieties of about 280 KÄ . Mean WTP for wine variety conservation was found to be about 130 KÄ , while WTP for conserving hop varieties was estimated at about 80 KÄ . Mean WTP values among respondents from South Moravia for crop conservation programs were found to be between about three and four times higher than for the general Czech population. We further examine the impact of observed preference heterogeneity for several respondent-specific characteristics on WTP for the conservation of the three crops. In total, the Czech adult population was estimated to have an aggregate WTP of at least two billion KÄ for additional fruit tree conservation over next ten years, about one billion KÄ for the conservation of additional wine varieties, and ~580 million KÄ for the conservation of additional hop varieties, and these values increase by 31–112 percent if the estimated benefits for the maximum number of varieties as offered in our design are added), revealing the previously unmeasured social welfare benefits of these activities. The estimated benefits of specific crop conservation are an important contribution to the valuation of these historic Czech resources, as crop varieties conserved now provide not only option and bequest values but may also be more resistant to biotic stresses (such as pests and diseases) as well as expected adverse weather extremes, providing the potential to help adapt Czech agriculture to future shocks.
    Keywords: Crop diversity; plant genetic resources for food and agriculture; discrete choice experiments; mixed logit; willingness to pay; consumer preferences
    JEL: Q18 Q51 Q57
    Date: 2020–08
  2. By: Wolff, Stefanie (E.ON Energy Research Center, Future Energy Consumer Needs and Behavior (FCN)); Madlener, Reinhard (E.ON Energy Research Center, Future Energy Consumer Needs and Behavior (FCN))
    Abstract: In this paper, we spatially map the willingness to pay for e-vehicle charging options according to the availability of public charging spots. We combine a Discrete Choice Experiment on charging preferences with a data set of public charging spots. Our results show spatial heterogeneity, i.e. respondents’ choices depend on the quantity of public charging spots available to them. Non-availability of public charging spots in the vicinity has a larger effect on the choice probability than 1, 2, or 3 charging spots have. This could be evidence for charging infrastructure awareness. For the charging locations, we find marked spatial heterogeneity in the willingness to pay subject to the number of available public charging spots. The interaction of charging location with the number of public charging spots reveals a strong preference for charging at home rather than at work or charging on the road. However, with every additional public charging spot, respondents are more likely to charge away from home. This holds until the number of charging spots has reached a tipping point at which respondents become indifferent between home and work charging. When the tipping point is exceeded, respondents rather charge at work than at home. Thus, with increasing numbers of charging spots, public chargers near home are less relevant than those near work. Eventually, public chargers away from home become more attractive. Also, with increasing numbers of charging spots our results reveal a fivefold greater willingness to pay for reducing waiting time (for a charging spot to become available) than for accelerating charging speed. Thus, charging point operators could surcharge by implementing a booking scheme than by implementing fast-charging. From the findings, we derive further implications for charging infrastructure policy, business models, and infrastructure planning, e.g. regarding the expected break-even points for rolling out charging infrastructure and the provision of green energy.
    Keywords: Electric mobility charging behavior; Charging spot awareness; Discrete Choice Experiment; Econometric modeling; Willingness to pay; Germany
    JEL: C25 D12 M38 Q58 R40
    Date: 2020–06
  3. By: Guillaume Monchambert (LAET - Laboratoire Aménagement Économie Transports - UL2 - Université Lumière - Lyon 2 - ENTPE - École Nationale des Travaux Publics de l'État - CNRS - Centre National de la Recherche Scientifique, UL2 - Université Lumière - Lyon 2, Université de Lyon)
    Abstract: Long-distance carpooling is an emerging mode in France and Europe, but little is known about monetary values of this mode attributes in transport economics. We conducted a discrete choice experiment to identify and measure the values of attributes of long-distance transport modes for a trip as a driver and as a passenger, with a special focus on carpooling. Around 1.700 French individuals have been surveyed. We use discrete mixed logit models to estimate the probability of mode choice. We find that the value of travel time for a driver who carpools is on average 13% higher than the value of travel time when driving alone in his/her car. The average value of travel time for a carpool trip as passenger is around 26 euros per hour, 60% higher than for a train trip and 20% higher than for a bus trip. Moreover, our study confirms a strong preference for driving solo over taking carpoolers in one's car. We also show that individuals traveling as carpool passenger incur a "discomfort" cost of on average 4.5 euros per extra passenger in the same vehicle. Finally, we identify robust socioeconomic effects affecting the probability of carpooling, especially gender effects. When they drive a car, females are less likely to carpool than male, but they prefer to carpool two passengers over only one passenger.
    Keywords: Value of time,Long-distance,Carpooling,Discrete choice experiment
    Date: 2020
  4. By: Benjamin Dovey; Herbert Ntuli
    Abstract: This paper reports a study done on Cape Town consumers, with the aim to understand how their purchasing decisions are shaped with respect to GMOs. A choice experiment approach was used to examine consumer preferences for biotechnology products in the food market. Four models were run to analyse the data, i.e., the conditional logit and random parameters logit models, with and without iteration terms. The results revealed a large status quo bias, indicating that the majority of consumers were reluctant to consume foods that incorporated certain technologies in their production process. The paper concludes by stating that producers need to align the development of GMOs to better meet consumers desires rather than suppliers. Finally, the paper takes an understanding that to increase the acceptance of GMOs, a larger portion of the population needs to be educated better about this technology.
    Keywords: GMOs, biotechnology, choice experiment, willingness to pay, consumer preferences
    JEL: P46 Q2 Q18 Q51 Q55
    Date: 2020–07
  5. By: Hongzhang Shao; Anton J. Kleywegt
    Abstract: In classical optimization problems under discrete choice models, the decision maker can control at most one attribute for each choice. Usually, this attribute is the price of a product. However, price is not the only attribute that can impact both the product's market share and profit margin. This paper studies a generalization of these problems, which allows multiple attributes to be controlled for each choice of product. Past research showed that we can solve the classical optimization problems efficiently, using methods like the market share reformulation. However, these methods do not apply to the generalized problems. We showed that such problems are still tractable, and can be solved efficiently as conic programs. This argument applies to optimization problems under multinomial logit models, and applies to problems under nested logit models when certain assumptions hold. These generalized problems are very common in retail and transportation industries. This paper allows practitioners to solve such problems directly using commercial solvers.
    Date: 2020–07
  6. By: Cai, Zhengyu
    Abstract: This study investigates why the strong form of the spatial equilibrium is weakly supported in the literature. Using a discrete choice model, it shows that the strong form of the spatial equilibrium is rarely observed because workers are imperfectly mobile from the perspective of researchers. Incorporating the discrete choice model, a Markov chain is used to model the spatial dynamics of the population distribution. For a given location choice set, the population distribution is shown to converge to a unique spatial steady state. Microdata from the American Community Survey show that the model assumption is reasonable and support the model predictions.
    Keywords: imperfect mobility,heterogeneity,spatial steady state,discrete choice model,Markov chain analysis
    JEL: J61 R10 C25 C15 C44
    Date: 2020
  7. By: Georgios Gerasimou
    Abstract: We study two random non-forced choice models that explain behavioural patterns suggesting that the no-choice outside option is often selected when people find it hard to decide between the market alternatives available to them, even when these are few and desirable. The decision-conflict logit extends the logit model with an outside option by assigning a menu-dependent value to that option. This value captures the degree of complexity/decision difficulty at the relevant menu and allows for the choice probability of the outside option to either increase or decrease when the menu is expanded, depending on how many as well as on how attractive options are added to it. The multicriteria logit is a special case of this model and introduces multiple utility functions that jointly predict behaviour in a multiplicative-logit way. Every multicriteria logit admits a tractable discrete-choice formulation.
    Date: 2020–08
  8. By: Ruben Loaiza-Maya; Didier Nibbering
    Abstract: The multinomial probit model is a popular tool for analyzing choice behaviour as it allows for correlation between choice alternatives. Because current model specifications employ a full covariance matrix of the latent utilities for the choice alternatives, they are not scalable to a large number of choice alternatives. This paper proposes a factor structure on the covariance matrix, which makes the model scalable to large choice sets. The main challenge in estimating this structure is that the model parameters require identifying restrictions. We identify the parameters by a trace-restriction on the covariance matrix, which is imposed through a reparametrization of the factor structure. We specify interpretable prior distributions on the model parameters and develop an MCMC sampler for parameter estimation. The proposed approach substantially improves performance in large choice sets relative to existing multinomial probit specifications. Applications to purchase data show the economic importance of including a large number of choice alternatives in consumer choice analysis.
    Date: 2020–07
  9. By: Monia Saïdi (CESAER - Centre d'Economie et de Sociologie Rurales Appliquées à l'Agriculture et aux Espaces Ruraux - AgroSup Dijon - Institut National Supérieur des Sciences Agronomiques, de l'Alimentation et de l'Environnement - INRAE - Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement); Jean-Sauveur Ay (CESAER - Centre d'Economie et de Sociologie Rurales Appliquées à l'Agriculture et aux Espaces Ruraux - AgroSup Dijon - Institut National Supérieur des Sciences Agronomiques, de l'Alimentation et de l'Environnement - INRAE - Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement); Stephan Marette (ECO-PUB - Economie Publique - AgroParisTech - Université Paris-Saclay - INRAE - Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement); Christophe Martin (CSGA - Centre des Sciences du Goût et de l'Alimentation [Dijon] - UB - Université de Bourgogne - AgroSup Dijon - Institut National Supérieur des Sciences Agronomiques, de l'Alimentation et de l'Environnement - CNRS - Centre National de la Recherche Scientifique - INRAE - Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement)
    Abstract: This article presents a new experimental protocol for estimating consumers' willingness-to-pay (WTP) for products involved in a reshuffle of geographical indications (GIs), e.g., a change of hierarchical levels within a restricted area. Although the collective reputation of a given GI depends on its temporal stability, reshuffling a GI area could make it better aligned with product quality or consumers' perception. We first provide a simple theoretical model in which consumers put a negative value on within-GI quality variance, thereby showing that reshuffling the GI designation scheme may increase WTP without any change in product quality. Using the experimental protocol, we evaluate consumer perceptions of different reshuffling scenarios for the vineyards of Marsannay, Burgundy, France. The results reveal a significant increase in WTP for the current distribution of products' quality. Elicited WTP values are then used to simulate the optimal GI reshuffle.
    Keywords: Willingness To Pay,Geographical indication,Wine,Experimental economics,wine appellation,quality signal,public policy JEL Classification: Q18,Q28,L66,Consentement a payer,Vin,Indication Géographique Protégée IGP
    Date: 2020–02
  10. By: Denuit, Michel; Mesfoui, Mhamed; Trufin, Julien
    Date: 2019–01–01
  11. By: Oz Shy
    Abstract: Using machine learning techniques applied to consumer diary survey data, the author of this working paper examines methods for studying consumer payment choice. These techniques, especially when paired with regression analyses, provide useful information for understanding and predicting the payment choices consumers make.
    Keywords: studying consumer payment choice; point of sale; statistical learning; machine learning
    JEL: C19 E42
    Date: 2020–06–23
  12. By: Leo Sleuwaegen (MSI, KU Leuven , Belgium, Naamsestraat 69, 3000 B Leuven); Peter M. Smith (MSI, KU Leuven , Belgium, Naamsestraat 69, 3000 B Leuven)
    Abstract: The decision to serve foreign markets through exports or foreign direct investment (FDI) has been studied within proximity-concentration models of location, mainly in the context of trade in goods. This paper adapts these models to account for the specific nature of services that are traded across borders. We show how services can be characterized by a bundle of attributes that collectively describe the service. These attributes are then tested to show how they affect the choice between exports and FDI using service-level data for firms in Belgium selling services abroad. Three different types of characteristic are shown to affect the export versus FDI decision: intangibility, the searchexperience-credence framework and the requirement for either the supplier or the client to physically move to the point of production.
    Keywords: Services; International Trade; Exports; Foreign Direct Investment; Service characteristics.
    JEL: F14 M16 L8
    Date: 2020–07
  13. By: Kevin Ducbao Tran
    Abstract: In online commerce, obfuscation strategies by sellers are hypothesized to mislead consumers to their detriment and to the profit of sellers. One such obfuscation strategy is partitioned pricing in which the price is split into a base price and add-on fees. While empirical evidence suggests that partitioned pricing affects consumer decisions through salience effects, its consumer welfare consequences are largely unexplored. Therefore, I provide a quantification of the welfare impact of the behavioral response to partitioned pricing. To do so, I derive a discrete choice model that jointly allows for differences in the reaction to marginal changes in add-on fees and the base price as well as a discontinuous effect of a zero fee. The model is based on a framework on limited attention and I estimate it using web scraped data of posted price transactions on eBay Germany. My results suggest under-reaction to marginal changes in the shipping fee, consistent with previous results in the literature. However, I also document a discontinuous positive effect of free shipping on consumer demand, which is novel to the literature. The combined impact of these effects on consumer welfare is less than six percent of consumer surplus. The welfare impact is attenuated because the maximum shipping fee on eBay is capped and the free shipping effect partly counteracts the under-reaction to shipping fees in expectation.
    Keywords: Partitioned pricing, limited attention, consumer welfare, shipping fees, eBay
    JEL: D12 D60 D83 L11
    Date: 2020
  14. By: Eric Bahel (Virginia Polytechnic Institute and State University, Blacksburg, VA); Yves Sprumont (Université de Montréal and CIREQ)
    Abstract: We describe the class of strategy-proof mechanisms for choosing sets of objects when preferences are additive and monotonic.
    JEL: D71
    Date: 2020–05
  15. By: Shun Katsuki (School of Economics and Management, Kochi University of Technology); Yoichi Hizen (School of Economics and Management, Kochi University of Technology)
    Abstract: Does voting solve intergenerational sustainability dilemma? Do voting rules matter for inducing people to collectively select a sustainable alternative that leaves more resources for future generations? To answer these questions, we conduct a laboratory experiment with human subjects in the framework of intergenerational sustainability dilemma game, in which the own-payoff maximizing choice by the current generation decreases the size of resource left for the subsequent generations. The choice is made by voting among the members of each generation, and we compare three voting rules, ordinary voting, whereby each person has one vote, proxy voting, whereby a part of people are given an extra vote on behalf of the subsequent generations, and two-ballot voting, whereby all people are given an extra vote. We observe that proxy voting and two-ballot voting improve the frequency of sustainable choice in comparison with ordinary voting, but the frequency is still low. This result implies that having people vote individually hardly achieves sustainable choices by successive generations even if the rules of voting are elaborated to some extent.
    Keywords: proxy vote, intergenerational sustainability dilemma, future generation, laboratory experiment
    Date: 2020–07
  16. By: Li,Shanjun; Zhu,Xianglei; Ma,Yiding; Zhang,Fan; Zhou,Hui
    Abstract: To promote the development and diffusion of electric vehicles, central and local governments in many countries have adopted various incentive programs. This study examines the policy and market drivers behind the rapid development of the electric vehicle market in China, by far the largest one in the world. The analysis is based on the most comprehensive data on electric vehicle sales, local and central government incentive programs, and charging stations in 150 cities from 2015 to 2018. The study addresses the potential endogeneity of key variables, such as local policies and charging infrastructure, using the border regression design and instrumental variable method. The analysis shows that central and local subsidies accounted for over half of the electric vehicles sold during the data period. Investment in charging infrastructure is much more cost-effective than consumer purchase subsidies. In addition, the policy that merely provided electric vehicles a distinctively license plate was strikingly effective. These findings demonstrate the varying efficacy across policy instruments and highlight the critical role of government in promoting fuel-saving technologies.
    Keywords: Energy and Environment,Energy Demand,Energy and Mining,Multi Modal Transport,Ports&Waterways,Energy Policies&Economics,Transport Services
    Date: 2020–08–13
  17. By: Sean Horan (Université de Montréal and CIREQ); Yves Sprumont (Université de Montréal and CIREQ)
    Abstract: We propose a class of decisive collective choice rules that rely on an exogenous linear ordering to partition the majority relation into two acyclic relations. The first relation is used to obtain a shortlist of the feasible alternatives while the second is used to make a final choice. In combination with faithfulness to the underlying majority relation, rules in this class are characterized by two desirable rationality properties: Sen’s expansion consistency and a version of Manzini and Mariotti’s weak WARP. The rules also satisfy natural adaptations of Arrow’s independence of irrelevant alternatives and May’s positive responsiveness.
    Keywords: majority rule, decisiveness, IIA, monotonicity, rational shortlist methods
    JEL: D71 D72
    Date: 2020–05

This nep-dcm issue is ©2020 by Edoardo Marcucci. It is provided as is without any express or implied warranty. It may be freely redistributed in whole or in part for any purpose. If distributed in part, please include this notice.
General information on the NEP project can be found at For comments please write to the director of NEP, Marco Novarese at <>. Put “NEP” in the subject, otherwise your mail may be rejected.
NEP’s infrastructure is sponsored by the School of Economics and Finance of Massey University in New Zealand.