Skip to main content
Disciplines
Automotive Business IT + Informatics Construction + Real Estate Electrical Engineering + Electronics Energy + Sustainability Insurance + Risk Finance + Banking Management + Leadership Marketing + Sales Mechanical Engineering + Materials
Events
DE
EN
Books
Journals
Topic Page
Marketing
Start single access now
Access for companies
EXTENDED SEARCH
Log in
Top
Customer Needs and Solutions
Published in: Customer Needs and Solutions 2/2015

01-06-2015 | Research Article

The Technological Conundrum: How Rapidly Advancing Technology Can Lead to Commoditization

Authors: Tat Y. Chan, Ravi Dhar, William Putsis

Published in: Customer Needs and Solutions | Issue 2/2015

Log in

Activate our intelligent search to find suitable subject content or patents.

search-config
loading …

Abstract

Much of the recent empirical IO research has been conducted in the context of relatively mature, stable (often consumer packaged goods) markets. In these markets, consumer preferences and competitive interaction are often characterized by relatively stable patterns over time. In contrast, modeling and estimating analogous patterns in rapidly changing technology-intensive industries, where brand preferences and technology evolve rapidly, can be challenging. Our primary objective in this research is to build on recent developments in econometric estimation and modeling to assess the time-varying shifts in consumer brand preferences in a rapidly changing technology industry. In doing so, we document an interesting technological conundrum—technological advancement can often lead to commoditization. In the Personal Computer (PC) market, this began with differentiation that was initially central to success in a market that has been increasingly commoditized over time. Methodologically, we adopt a factor-analytic approach to model and estimate the evolution of consumer brand preferences over time.
previous article JB Steenkamp on the Success of Marketing Scholarship in Netherlands
next article Base + Add-on Pricing in Extremely Competitive Categories

Dont have a licence yet? Then find out more about our products and how to get one now:

Springer Professional "Wirtschaft"

Online-Abonnement

Mit Springer Professional "Wirtschaft" erhalten Sie Zugriff auf:

  • über 67.000 Bücher
  • über 340 Zeitschriften

aus folgenden Fachgebieten:

  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Finance + Banking
  • Management + Führung
  • Marketing + Vertrieb
  • Versicherung + Risiko




Jetzt Wissensvorsprung sichern!

inform now

Springer Professional "Wirtschaft+Technik"

Online-Abonnement

Mit Springer Professional "Wirtschaft+Technik" erhalten Sie Zugriff auf:

  • über 102.000 Bücher
  • über 537 Zeitschriften

aus folgenden Fachgebieten:

  • Automobil + Motoren
  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Elektrotechnik + Elektronik
  • Energie + Nachhaltigkeit
  • Finance + Banking
  • Management + Führung
  • Marketing + Vertrieb
  • Maschinenbau + Werkstoffe
  • Versicherung + Risiko

Jetzt Wissensvorsprung sichern!

inform now
Footnotes
1
There are a number of other studies that also employ the state-space approach. For example, Van Heerde et al. [20] used a dynamic linear model (see [6]) to investigate how consumer preferences are impacted by product innovations. They found, for example, that innovations make brands more similar and decrease differentiation for existing brands. These results support our findings at a number of levels (we will provide details later in the paper).
 
2
Other empirical studies using the same approach include Carroll et al. [5], Chintagunta [7], and Erdem and Winer [12]. Based on panel data, Erdem [11] also models how state dependence components affect consumer preferences for the latent attributes.
 
3
Detailed discussions on the IDC data can be found in Bayus and Putsis [2] and Putsis and Bayus [16] as well as on the IDC web site www.​idc.​com. Excellent historical reviews of the personal computer industry are given in Langlois [15] and Steffens [19]. Perhaps the most comprehensive reference site for a chronology of the industry can be found at www.​islandnet.​com/​∼KPOLSSON/​comphist/​.
 
4
By the mid to late 1990s, the industry had matured significantly and was certainly a relatively stable market in the early part of the new millennium. Further, although we have data after 1994, IDC altered its data collection methodology in 1994 and there are important differences in reported sales and pricing pre and post 1994. Hence, it is difficult, if not impossible, to directly combine the two data sets in model estimation. We feel strongly that we should err on the side of being conservative with the data, using only data for which we have time-related consistency in the way the data was collected and measured.
 
5
Another major brand in our sample period is Apple. Consumers who switch from Wintel to Apple have to invest in manufacturer-specific technology adoption. Apple is very different from the other seven brands from a number of perspectives; hence, consumer’s choice of switching in this context can be very different from what is implied in our empirical model below. As a result, Apple is included in the “outside” option in the demand model below.
 
6
We will provide detailed definitions for frontier and older technologies later in the paper.
 
7
The outside option is addressed in Footnote 5 above and will be further discussed later.
 
8
We also note that to estimate full heterogeneity and time evolution in consumer preferences for x jt (i.e., through β) is a non-trivial task. Since consumer preferences for the x jt themselves are not the focus of, nor generally material to the current study, we limit our approach in the interest of parsimony and in keeping the estimation manageable and focused at the issues at hand.
 
9
The standard normal distribution assumption follows Elrod and Keane [10], which is necessary for model identification.
 
10
Suppose we have data on marketing efforts from upstream manufacturers (e.g. Intel’s “Intel Inside” campaign) that should only impact the change in the preference weights for the brand attributes, \( {\overline{c}}_t \), and data on marketing efforts from branded manufacturers, which should only have effect on the change in the brand attribute levels, L mt . It is possible to identify both changes from model estimation. In this study, however, we do not have these data.
 
11
To estimate both model specifications, further normalization restrictions are required. Details are below.
 
12
\( {\overline{c}}_t \) in Model 1 can be estimated for each attribute in each period, assuming that the number of attributes K is much smaller than the number of manufacturer brands M (7 in the study).
 
13
The mean value of these normalized brands in consumer utility will be captured by the intercept x jt in equation (1). The heterogeneity in brand values among these brands will be absorbed by demand shocks ξ jt .
 
14
386 and 486 were introduced into the market late in the years 1986 and 1989, respectively. There were very few PC products incorporating these two technologies in their introduction years. Hence, 286 and 386 were still the frontier technology in years 1986 and 1989, respectively.
 
15
Although τ it is interpreted as the preference for frontier technology, it can also be interpreted as the value loss of a technology once it is replaced by a faster and more powerful microprocessor technology, e.g., some software are designed for the new one but are incompatible with old ones.
 
16
For the simplicity of illustration, we still use L mt to represent L m in model 1, and \( {\overline{c}}_t \) to represent \( \overline{c} \) for model 2.
 
17
We test for price endogeneity employing a Hausman-Durbin-Wu test, assuming that we have valid instruments for prices (see discussion below). The null hypothesis that price is exogenous can be rejected at p = .05.
 
18
To calculate market share from sales, we need to know the potential market size for personal computers, which is not observed from the data. We do not want to impose the standard assumption in the literature that it is equal to the total population or the number of households, since consumers in our data also include firms, whose needs of PCs may vary significantly depending on their industries and business scale. Therefore, we assume that the potential market size in every year t is equal to the total PC sales in 1999, 4 years after our sample period ends. We test the impact of such assumption on estimation results by re-estimating our model assuming the potential market size is 50 % smaller and larger than this size. Our major results (e.g., time-varying brand preferences, price, and technology coefficients) remain qualitatively the same. Hence, we conclude that our key findings are robust to the potential market size assumption.
 
19
Note that IBM and Compaq are leaders in attribute 1. IBM was the first PC manufacturer using operating systems from Microsoft. Compaq was one of the first who pushed out new PC models incorporating Intel 386 microprocessor in 1987.
 
Literature
1.
Allenby GM, Lenk PJ (1995) Reassessing brand loyalty, price sensitivity, and merchandising effects on consumer brand choice. J Bus Econ Stat 13(3):281–289
2.
Bayus B, Putsis W (1999) Product proliferation: an empirical analysis of product line determinants and market outcomes. Mark Sci 18(2):137–153CrossRef
3.
Berry S, Levinsohn J, Pakes A (1995) Automobile prices in market equilibrium. Econometrica 63(4):841–890CrossRef
4.
Bresnahan TF, Greenstein S (1997) “Technological competition and the structure of the computer industry,” manuscript.
5.
Carroll JD, De Soete G, DeSarbo WS (1990) Two stochastic multidimensional choice models for marketing research. Decis Sci 21:337–356CrossRef
6.
Chib S, Greenberg E (1995) Hierarchical extensions of sur models with extensions to correlated serial errors and time-varying parameter models. J Econ 68:339–360CrossRef
7.
Chintagunta P (1994) Heterogeneous logit model implications for brand positioning. J Mark Res 31(May):304–311CrossRef
8.
Putsis W, Dhar R (2009) “Category expenditure and promotion: can private labels expand the pie?” unpublished manuscript, London Business School.
9.
Elrod T (1998) Choice map: inferring a product-market map from panel data. Mark Sci 7(1):21–40
10.
Elrod T, Keane M (1995) A factor-analytic probit model for representing the market structure in panel data. J Mark Res 32(1):1–16CrossRef
11.
Erdem T (1996) A dynamic analysis of market structure based on panel data. Mark Sci 15(4):359–378CrossRef
12.
Erdem T, Winer R (1999) Econometric modeling of competition: a multi-category choice-based mapping approach. J Econ 89(1–2):159–175
13.
Gordon BR (2009) A dynamic model of consumer replacement cycles in the PC processor industry. Mktg Sci 28(5):846–867
14.
Lachaab M, Ansari A, Jedidi K, Trabelsi A (2006) Modeling preference evolution in discrete choice models: a Bayesian state-space approach. Quant Mark Econ 4:57–81CrossRef
15.
Langlois R (1992) External economies and economic progress: the case of the microcomputer industry. Bus Hist Rev 66:1–50CrossRef
16.
Putsis W, Bayus B (2001) An empirical analysis of firm product line decisions. J Mark Res 38(1):110–118CrossRef
17.
Seetharaman PB (2004) Modeling multiple sources of state dependence in random utility models of brand choice: a distributed lag approach. Mark Sci 23(2):234–242CrossRef
18.
Smith RJ (1992) Non-nested tests for competing models estimated by generalized method of moments. Econometrica 60(4):973–980CrossRef
19.
Steffens J (1994) New games: strategic competition in the PC revolution. Pergamon Press, New York
20.
Van Heerde H, Mela CF, Manchanda P (2004) The dynamic effect of innovation on market structure. J Mark Res 41(2):166–183CrossRef
Metadata
Title
The Technological Conundrum: How Rapidly Advancing Technology Can Lead to Commoditization
Authors
Tat Y. Chan
Ravi Dhar
William Putsis
Publication date
01-06-2015
Publisher
Springer US
Published in
Customer Needs and Solutions / Issue 2/2015
Print ISSN: 2196-291X
Electronic ISSN: 2196-2928
DOI
https://doi.org/10.1007/s40547-015-0047-y

Other articles of this Issue 2/2015

Customer Needs and Solutions 2/2015 Go to the issue

Perspectives

Fair Wealth

Research Article

Wireless Carriers’ Exclusive Handset Arrangements: an Empirical Look at the iPhone

Perspectives

JB Steenkamp on the Success of Marketing Scholarship in Netherlands

Research Article

Retaining Talented Salespeople

Research Article

Implementing Retail Category Management: a Model-Based Approach to Setting Optimal Markups

Research Article

Base + Add-on Pricing in Extremely Competitive Categories