The 2012 Workshop on Information Systems and Economics (WISE), Orlando, FL., December 2012.

Size: px

Start display at page:

Download "The 2012 Workshop on Information Systems and Economics (WISE), Orlando, FL., December 2012."

Alfred Harrell
6 years ago
Views:

Title Pricing strategy and technology choices: an empirical investigation of Everyday Low Price in the domestic US Airline sector Athor(s) Sin, R; Chellappa, RK Citation The 2012 Workshop on

1 Title Pricing strategy and technology choices: an empirical investigation of Everyday Low Price in the domestic US Airline sector Athor(s) Sin, R; Chellappa, RK Citation The 2012 Workshop on Information Systems and Economics (WISE), Orlando, FL., December Issed Date 2012 URL Rights This work is licensed nder a Creative Commons Attribtion- NonCommercial-NoDerivatives 4.0 International License.

2 Pricing Strategy and Technology Choices: An Empirical Investigation of Everyday Low Price in the Domestic US Airline Sector 1. Introdction Extended Abstract There is a rich literatre in economics on factors that govern airline prices. With approximately 50% of airline tickets sold online, there is now a renewed interest in investigating airline pricing particlarly amongst Information Systems (IS) researchers. While market transparency created by online travel agents (OTAs) is a motivation enogh to reexamine airline pricing, one missing piece calls for a thorogh empirical investigation: In all extant stdies (economics, marketing and IS), pricing by two major airlines, Sothwest and JetBle, has been ignored. Of particlar interest to IS researchers is the fact that these two airlines practice a niqe form of pricing called Everyday Low Price (EDLP) and have made certain technology-related choices that are distinct and different from previosly stdied carriers. We test extant theories of price discrimination in the new online context with both posted and transacted prices, inclding those of the two EDLP airlines. We find that the EDLP airlines demonstrate distinctly different pricing choices they are very particlar abot maintaining price consistency, while except in the case of advance prchase, they forgo other conventional opportnities to price discriminate. They also aggressively nderct competition when eqipped with certain cost advantages. Frther empirical investigation reveals why EDLP airlines forgo participation in OTAs that potentially offer a larger market. In particlar, while promising a larger marketplace, OTAs have the ability to reveal the lowest prices in the market; we find that EDLP prices are lowest in the market only 26% of the time in other words, consmers have 70% chances of obtaining a better deal from other non-low price airlines for any given ticket. We also show that EDLP prices, thogh not necessarily being the lowest in the market, do tend towards the lower end of the market spectrm. As a first step, we decompose a firm s pricing strategy into two dimensions, namely price consistency and price magnitde. The former captres the freqency of promotions in tickets for the HILO airlines and is a measre of the everyday aspect of their EDLP conterparts, while the latter allows s to examine the low price element of the pricing strategy. We then develop and estimate a hierarchical model that acconts for partial dependence among prices de to market- and airline-specific factors, sing pricing data for over 200,000 tickets offered online by the 14 largest carriers in 268 U.S. domestic air travel markets. In constrcting this model, we careflly control for potential endogeneity that may arise from market share-related measres. Frther, since or model incorporates a nmber of dichotomos variables at varios levels (e.g. advance prchase and Satrday stay-over, hb operation, rote distance), we exercised extreme cation in interpreting related effects and 1

3 drawing inferences from or statistical analysis. We also repeat all of or analyses with the transacted price data from DB1B (a 10% sample of all tickets sold by reporting carriers, provided by the Brea of Transportation Statistics) both as a robstness check of or original reslts and as a means to verify the extent to which the pricing strategies of airlines are absorbed by the market. While existing research typically investigates either offered prices or transacted prices, or work examines firm-specific strategies (and hence the respective price setting behaviors) and conclde with similar examination of prices at which tickets have actally been prchased. 2. Data Or data is collected from two primary sorces. First, we obtained prices and detailed descriptions of airline tickets from online travel agents and individal airlines websites in the third qarter of This raw data was gathered sing web-based spiders that we developed sing Crl, and later processed by a parser sing Perl and other database scripting langages. In addition to the set of forteen major U.S. carriers 1 and three online travel agents (Orbitz, Travelociy, and Expedia), a list of the top 500 U.S. domestic rotes was provided as inpt to the spiders. These markets represent over 86% of total domestic passenger enplanements (i.e. total nmber of travelers transported by air within the 50 states) in the U.S. as of The spiders were sent ot on a daily basis to collect prices and other attribtes of tickets reqiring one- to for-week advance prchases, inclding weekday as well as weekend departres and retrns. Or agents operated in parallel and sbmitted identical reservation reqests to all online travel agents and airlines websites simltaneosly in order to minimize price variations that may arise from the timing of ticket reqests. Consistent with prior research on airline pricing, we consider only coach class, nonrefndable, rond-trip tickets. Frther, to control for any price difference that may be attribted to differences in flight dration or the nmber of connections on any given rote, we restrict or attention to non-stop flights between an origin and a destination. Since non-stop flights were not available in 28 rotes, or data set is redced to 472 markets with 272,362 niqe tickets and final prices, inclding taxes and fees, offered by forteen largest domestic carriers 2. Second, we sed the Origin and Destination Srvey (DB1B) from the Brea of Transportation Statics (BTS) for the corresponding rotes and carriers in the second and third qarters of 2004 to collect prices on transacted tickets and compte the market shares of individal airlines in each origin-destination pair. DB1B is a 10% sample of all tickets sold by reporting carriers, inclding origin, destination and other itinerary details of passengers transported. This data contains sales from both online and offline channels, and is sed in or 1 The forteen major carriers are American, Alaska, Continental, Delta, United, US Airways, Sothwest, JetBle, Frontier, America West, AirTran, ATA, and Spirit. 2 Spirit Airlines had been dropped from or sbseqent analysis de to insfficient nmber of observations. 2

4 later analysis to check the robstness of or findings that are based on the online pricing data we gathered from the Internet. In addition, we sed the Air Carrier Statistics (Form 41 Traffic and 298C Smmary Data) and Air Carrier Financial Reports (Schedle P-12) provided by the BTS to assemble data on airlines operational details (e.g. cost per available seat-mile, aircraft types and sizes, freqency of flights, etc.), as well as information on the respective markets (e.g. origin-destination distance, hb information, etc.). By combining these varios sets of data, we yielded a complete profile of all relevant variables at the ticket level that allowed s to examine the effects of varios market- and firmspecific factors on airline pricing. We sbseqently eliminated rotes where the dominant carrier has over 90% market share. Or final data set consists of 209,558 observations from 268 markets. Table 3 reports the descriptive statistics of or data. 3. Models Or econometric estimation aims to provide an nderstanding of firms pricing strategies throgh two distinct measres. The first is price which is self-explanatory in its ability to describe pricing strategy. In the price model, a positive (negative) coefficient for an independent variable sggests that the variable is correlated with a higher (lower) ticket price. The second dependent variable is price variance commonly in IS literatre this variable has only been sed for stdying market level dispersion in prices across firms. However, the marketing literatre has shown how variance in prices of a single firm is in itself an exection of a corresponding pricing strategy. For example, the work by Hoch et al. (1994), Ho et al. (Ho, Tang et al.), and Shankar and Bolton (2004) se price variance at the firm-level to show some firms vary prices freqently while others do not; the theoretical reasoning behind creating mltiple price points is that these different prices can appeal to different segments of consmers, allowing the firm to extract more srpls. Ths an examination of this dependent variable is critical to or nderstanding of the extent to which airlines like Sothwest and JetBle prse discrimination compared with the other major carriers. Model 1: Price level Consistent with extant literatre on airline pricing, we employ log-transformation for most of or explanatory variables to captre their declining marginal effects on prices. The variables that are inclded in the model withot any transformation are the dmmy variables, market share ( RTshare ), and the corresponding Herfindahl index( RTherf ) (Borenstein 1989). Level 1 (ticket-level) model: ln( price ) = a + a BUS + a DD7 + a DD14 + a DD21 + e ikmc e ikmc 0kmc 1kmc ikmc 2kmc ikmc 3kmc ikmc 4kmc ikmc ikmc 2 ( s ) In this model, the dependent variable in market (rote) m in a given ticket category ( c ). price ikmc denotes the price of ticket i offered by carrier k (1) 3

5 Level 2 (airline- and market-level) model: a0 km = b0 + g01edlpk + g02 ln( freq ) + g03hb km km + g04rtsharekm + g05rtherfm + g06shorthalm + g07 ln( CASM ) + g08 ln k ( EQUIPsizekm ) + g EDLP shorthal + g EDLP hb a = g + d ( ) ( ) 09 k m 10 k km 00k 00m 0km 1-4km 1-4km 1-4km k 00k 00m 0km ( j ) ( t ) ( y ) EDLP The fll model (after rearranging terms and renaming the coefficients): ( ) ln price = a + b EDLP + b BUS + b DD7 + b DD14 + b DD21 where ikmc 1 k 2 ikmc 3 ikmc 4 ikmc 5 ikmc + b6ln( freqkm ) + b7hbkm + b8rtsharekm + b9rtherfm + b10shorthalm + b11 ln( CASMk ) + b12 ln( EQUIPsizekm) + b13 ( EDLPk BUSikmc ) + b14 ( EDLPk DD7ikmc ) + b15 ( EDLPk DD14ikmc ) + b ( EDLP DD21 ) ( EDLP shorthal ) ( EDLP hb ) + b + b + e 16 k ikmc 17 k m 18 k km ikm a = b (4) 0 00k 00m 0km Eqation (3) is the basic model to be estimated. Interactions between the EDLP identifier and varios ticket categories (Satrday night stay-over and advance prchase periods) and market characteristics (short-hal and hb) are inclded to captre any potential differences in the pricing approach of EDLP verss other major carriers de to differences in their segmentation approach and operational/cost differences in varios types of markets. b 0 represents the overall intercept; 00k and 00m are the random carrier and rote effects, respectively. 0km is the random interaction effect. Finally, e ikm is the white-noise error particlar to the individal observation. Model 2: Price variance where CV = a + b EDLP + b BUS + b DD7 + b DD14 + b DD21 kmc 1 k 2 kmc 3 kmc 4 kmc 5 kmc + b freq + b hb + b RTshare + b RTherf 6 km 7 km 8 km 9 m + b shorthal + b CASM + b EQUIPsize 10 m 11 k 12 km + b13 ( EDLPk BUSkm) + b14 ( EDLPk DD7km) + b15 ( EDLPk DD14km) + b ( EDLP DD21 ) + b ( EDLP shorthal ) b ( EDLP hb ) 16 k km e k m 18 k km km a = b + (6) 0 0m The dependent variable in Model 2 is the coefficient of variation of prices, which is measred at the carrier-rote-ticket category level and is constrcted from the set of tickets written by an airline ( k ) in a particlar rote ( m ) for a given ticket category ( ) 4. Discssion Sothwest Airlines completely eschews participation in any of the OTAs while JetBle does the same while offering some selected fares and participates partially in one OTA. Both airlines c. (2) (3) (5) 4

6 primarily rely on selling tickets throgh their own web sites; in particlar, Sothwest had created its own GDSs and has generally adopted a postre different from other airlines. We primarily seek to nderstand the jdiciosness of this choice throgh or empirical analyses particlarly since low price airlines shold benefit from flanting and even being compared to other airlines. Or reslts bear ot the fll story. The sorting and varios other comparison mechanisms offered by the OTAs provide significant improvements in prodct and price transparency, allowing the consmer to expeditiosly determine the lowest price in the marketplace. Sch an impact of technology cannot be flly assessed by the standard econometric specification, which concerns the average price of an airline with relations to its competitors. Therefore, in addition to the above estimations, we condct two empirical analyses: a simple Logit model to assess the probability of these two EDLP airlines offering the lowest fare in the marketplace, and a model that determines how far they sally stray from the market minimm. It is also interesting to note at this point that American Airlines (AA) have made some very crios moves regarding their participation at OTAs. AA opted ot from Orbitz in late 2010, which reslted in Expedia to pll their American offers from their websites. Shortly afterwards (in April 2011), American filed a lawsit against Orbitz for exclsionary and anticompetitive bsiness practices while at the same time reaching an agreement with Expedia to have their offers once again displayed by the OTA. While the indstry sspected that AA withdrew from OTAs primarily to ct distribtion cost, and later staged a retrn becase they realized that the savings are not worth losing ot large marketplaces, American seems to be reformlating their pricing and technology choices. In fact, at the same time AA accsed Orbitz of anticompetitive condct, they also annonced the development of their own distribtion channel, known as the Direct Connect. Essentially, American is demanding OTAs to retrieve flight information directly from this new system rather than throgh the GDSs. Sch moves perhaps indicate that American realizes both the potential offered by the OTAs and the threats of transparency associated with sch technologies; ths in response, they develop a new channel that allow for the best of both worlds throgh selectively listing their most competitive offers on OTAs while presenting a different set of choices to consmers who visit their own website. References Bell, D. R. and J. M. Lattin Shopping Behavior and Consmer Preference for Store Price Format: Why "Large Basket" Shoppers Prefer EDLP. Marketing Science 17(1) Berry, S Airport Presence as Prodct Differentiation. The American Economic Review 80(2) Berry, S., M. Carnall and P. T. Spiller Airline Hbs: Costs, Markps and the Implications of Cstomer Heterogeneity. NBER Working Paper #5561. Borenstein, S Hbs and High Fares: Dominance and Market Power in the U.S. Airline Indstry. The RAND Jornal of Economics 20(3) Borenstein, S The Dominant-Firm Advantage in Mlti-Prodct Indstries: Evidence from the U.S. Airlines. Qarterly Jornal of Economics 106(November). 5

7 Borenstein, S. and N. L. Rose Competition and Price Dispersion in the U.S. Airline Indstry. The Jornal of Political Economy 102(4) Chellappa, R. K., R. G. Sin and S. Siddarth Price-Formats as a Sorce of Price Dispersion: A Stdy of Online and Offline Prices in the Domestic US Airline Markets Information Systems Research 22(1). Clemons, E. K., I. Hann and L. M. Hitt Price Dispersion and Differentiation in Online Travel: An Empirical Investigation. Management Science 48(4) Dana, J. D Advance-Prchase Disconts and Price Discrimination in Competitive Markets. The Jornal of Political Economy 106(2) Dliba, K. A., R. J. Kaffman and H. C. Lcas Appropriating Vale from Compterized Reservation System Ownership in the Airline Indstry. Organization Science 12(6) Evans, W. N. and I. N. Kessides Living by the "Golden Rle": Mltimarket Contact in the U.S. Airline Indstry. The Qarterly Jornal of Economics 109(2) Gale, I. L. and T. J. Holmes Advance-Prchase Disconts and Monopoly Allocation of Capacity. The American Economic Review 83(1) Gimeno, J. and E. Jeong Mltimarket Contact: Meaning and Measrement at Mltiple Levels of Analysis. Mltinit Organization and Mltimarket Strategy. J. A. C. Bam and H. Greve. JAI Press, London. Granados, N., A. Gpta and R. J. Kaffman The Impact of IT on Market Information and Transparency: A Unified Theoretical Framework. Jornal of the Association for Information Systems March Granados, N., A. Gpta and R. J. Kaffman Information Transparency in Bsiness-to-Consmer Markets: Concepts, Framework, and Research Agenda. Information Systems Research 21(2) Granados, N., R. J. Kaffman and B. King How Has Electronic Travel Distribtion Been Transformed? A Test of the Theory of Newly Vlnerable Markets. Jornal of Management Information Systems 25(2) Morrison, S. A Actal, Adjacent, and Potential Competition: Estimating the Fll Effect of Sothwest Airlines. Jornal of Transport Economics and Policy Neels, K Congestion Pricing and the Economic Reglation of Airports. Airports in the 21st Centry: Proceedings of a Conference. J. M. Rakas, G. W. Blomme and G. Gosling Transportation Research Board, The National Academies. Washington, D.C Nevo, A Measring Market Power in the Ready-to-eat Cereal Indstry. Econometrica 69(2) Stavins, J Price Discrimination in the Airline Market: The Effect of Market Concentration. Review of Economics and Statistics 83(1) Venkatesan, R., K. Mehta and R. Bapna Understanding the conflence of retailer characteristics, market characteristics and online pricing strategies. Decision Spport Systems 42(3)

USING PERSONAL ASSISTANCE AS A WORKPLACE SUPPORT

1 USING PERSONAL ASSISTANCE AS A WORKPLACE SUPPORT by: Ed Trner For many people with significant disabilities, the se of Personal Assistance Services (PAS) as a workplace spport is as essential to sccessfl