nep-dcm New Economics Papers
on Discrete Choice Models
Issue of 2014‒01‒10
four papers chosen by
Edoardo Marcucci
Universita' di Roma Tre

  1. Estimation of switching costs and network effects in mobile telecommunications in Poland By Czajkowski, Mikołaj; Sobolewski, Maciej
  3. "Intensive Margins, Extensive Margins, and the Spousal Allowances in the Japanese System of Personal Income Taxes: A Discrete Choice Analysis" By Shun-ichiro Bessho; Masayoshi Hayashi
  4. Dynamics of broadbands demand: Substitution or complementarity between fixed and mobile technologies? An application to the Portuguese case By Silva, Rita Filipe; Proença, Isabel; Vareda, João

  1. By: Czajkowski, Mikołaj; Sobolewski, Maciej
    Abstract: In this paper we utilize discrete choice experiment method to identify and measure switching costs and network effects in mobile telephony in Poland. Based on hypothetical choices consumers make we construct a conditional random parameters multinomial logit model to analyze their preferences. In our choice design we explicitly account for status quo inertia, number portability, operator brand, network distribution of most frequently called parties and price of on-net and off-net calls. Stated preference approach allows us to calculate marginal rates of substitution and hence implicit prices of the non-price attributes used to describe choices and switching behavior. Results of our study indicate that although choices of mobile operators are largely driven by price of calls, switching costs and network effects have and strong impact on utility of subscribers. In particular users assign positive value to their mobile phone number and the size of family and friends group in the same network. The monetary value of phone number is significantly higher among individual entrepreneurs then residential subscribers. In our model switching behavior is not discouraged by brand loyalty which turned out to be insignificant. Instead subscribers follow status quo inertia which reflects uncertainty associated with new operator. Therefore we conclude that despite introduction of mobile number portability, switching costs continue to be an important issue in telecommunications markets. On recommendations level, we argue that regulatory and competition policies should continue to reduce uncertainty associated with changing operator by ensuring service and platform compatibility and reducing tariff complexity. In light of our results we recommend tariffs to be non-discriminatory so that operators are unable to utilize network effects in a way which discourages switching behavior. --
    Keywords: Switching costs,network effects,mobile telecommunications,mobile number portability,brand valuation,stated preference methods,non-market valuation methods,choice experiment,multinomial conditional logit model,random parameters model
    JEL: L1 L86 O3
    Date: 2013
  2. By: Aygul Ozbafli (JDINT'L Department of Economics Queen's University, Canada); Glenn Jenkins (Department of Economics, Queen's University, Canada, Eastern Mediterranean University, Mersin 10, TURKEY)
    Abstract: This research examines households’ willingness to pay (WTP) for an improved electricity service. Households’ stated WTP is estimated using the choice experiment method (CE). The data used in the estimations came from 350 in-person interviews conducted during the period 5–22 August 2008 in North Cyprus. Compensating variation (CV) estimates for a zero-outage scenario are calculated using the parameter estimates from the mixed logit (ML) model; these are 6.65 YTL (Turkish lira) per month (5.66 USD) for summer and 25.83 YTL per month (21.97 USD) for winter. In order to avoid the cost of outages, households are willing to incur a 3.6% and a 13.9% increase in their monthly electricity bill for summer and winter, respectively. The WTP per hour unserved is 0.28 YTL (0.24 USD) for summer, and 1.08 YTL (0.92 USD) for winter. A preliminary cost–benefit analysis indicates that the annualized economic benefits are approximately 16.3 million USD for the residential sector, and justify an investment in additional generation capacity of approximately 120 MW.
    Keywords: Willingness to pay; choice experiment; electricity; outages; reliability
    JEL: D12 D61 L94 L98 Q41
    Date: 2013–11
  3. By: Shun-ichiro Bessho (Faculty of Economics, Keio University,); Masayoshi Hayashi (Faculty of Economics, The University of Tokyo)
    Abstract:    This study explores the effects of the spousal allowances in the Japanese system of personal income taxes, taking advantage of the micro-simulation method based on the discrete choice model of labor supply. Our simulations show that the complete abolishment of the spousal allowances would increase the female labor supply by 1.6% for all wives, and by .1% for wives that are supposed to be under a large influence of the allowances. We also examine the reform in the allowances which leads to a decrease in the female labor supply. We argue that these results are due to our explicit consideration of the fixed cost of labor market participations, which has been ignored in the previous Japanese studies.
    Date: 2013–12
  4. By: Silva, Rita Filipe; Proença, Isabel; Vareda, João
    Abstract: The development of the broadband market is a key aspect of the economic and social growth of a country. However, despite the importance and the development of broadband market in Portugal in recent years, especially with the explosion of the number of mobile broadband accesses, the studies for the Portuguese case are rare. The present paper seeks to contribute to the discussion about the definition of the broadband market in Portugal, specifically studying the demand for broadband Internet and measuring the determinants that explain the use of each of the technologies available to provide broadband access, with emphasis on the differences between fixed and mobile accesses. Demand broadband functions were estimated using nested logit and multinomial discrete choice model. The primary source of information was ANACOM's Electronic Communications Services Consumption Survey, complemented with price information regarding the offers available in the market. The estimations obtained for the elasticities point out the probable existence of substitution between ADSL and cable and between these fixed broadband technologies and the mobile broadband. However, the inverse relation is not statistically significant, the demand for mobile broadband isn't constrained by the price of ADSL or of cable, which may reveal the existence of asymmetric substitution between fixed and mobile broadband accesses. These results have implications in the definition of the broadband market in Portugal which will be discussed. --
    Keywords: Broadband,Market Definition,Demand,Regulation,Fixed-Mobile Substitution
    JEL: L51 L96
    Date: 2013

This nep-dcm issue is ©2014 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.