Abstract
We develop a market-based paradigm to value the enhancement or addition of features to a product. We define the market value of a product or feature enhancement as the change in the equilibrium profits that would prevail with and without the enhancement. In order to compute changes in equilibrium profits, a valid demand system must be constructed to value the feature. The demand system must be supplemented by information on competitive offerings and cost. In many situations, demand data is either not available or not informative with respect to demand for a product feature. Conjoint methods can be used to construct the demand system via a set of designed survey-based experiments. We illustrate our methods using data on the demand for digital cameras and demonstrate how the profits-based metric provides very different answers than the standard welfare or Willingness-To-Pay calculations.
Similar content being viewed by others
Notes
Valuation of product features is also critical in patent disputes. In particular, damages from patent infringement are typically assessed as either royalty payments or lost profits. In either case, a valuation of the patented feature in terms of incremental profits should be at the heart of any valid damages calculation. In a companion paper, we consider the problem of patent damage calculations and provide explicit methods for using conjoint data to undertake these calculations (Allenby et al. 2014).
In equilibrium after a feature introduction, it is possible that there may be quantity adjustments which may change the marginal cost of production. It would be a simple matter to add a specification in which cost changes as a function of quantity produced. It should be noted that this is yet another way in which the failure to account for equilibrium adjustments may bias traditional approaches to feature valuation such as Willingness To Pay.
That is, if price takes on K values, \(p_{1},\dots ,p_{K}\), then many conjoint investigators include K−1 dummy variables for each of the values. This makes the market demand a non-continuous function and can create a situation in which there does not exist an equilibrium price. Existence of pure strategy equilibria requires (at a minimum) a continuous best-response function. If utility is a discontinuous function of price, then there can be discontinuities in the best response functions. As our proposed method requires equilibrium calculations, we do not use a dummy variable coding. Nonlinearities in the utility function with respect to price can be handled via non-linear continuous functions, if desired.
There is no guarantee that a Nash equilibrium exists for heterogeneous logit demand.
Again, we do not have an aggregate demand shock in the model. We think of the firm problem as setting prices given the observed characteristics and prices of all products in the marketplace. There is no sense in which firms are setting prices as a function of some unobserved characteristic as this is explicitly ruled out by the nature of the conjoint randomized experiment.
We do not include a market wide shock to demand as we are not trying to build an empirical model of market shares. We are trying to approximate the firm problem. In a conjoint setting, we abstract from the problem of omitted characteristics as the products we use in our market simulators are defined only in terms of known and observable characteristics. Thus, the standard interpretation of the market wide shock is not applicable here. Another interpretation is that the market wide shock represents some sort of marketing action by the firms (e.g. advertising). Here we are directly solving the firm pricing problem holding fixed any other marketing actions. This means that the second interpretation of the market wide shock as stemming from some unobservable firm action is not applicable here.
In the conjoint literature, this is often termed the “choice share.”
It should be noted that the prior outlined here will have a zero probability of a price coefficient which is ≥0. This is not true for more ad hoc methods such as the method of “tie-breaking” used in Sawtooth Software.
This study was part of a wave of four other very similar conjoint studies on digital cameras each with the same screening criteria. For all studies in the wave, 16,185 invitations were sent to panelists, 6,384 responded. Of those who responded to the invitation, 2,818 passed screening and of those passing screening 2,503 completed the questionnaire. Thus, the overall completion rate is 89 per cent which is good by survey standards.
The numbers displayed in the table are posterior means.
The profits per unit sold are $172-75 and 35 per cent of this is $33.
However, survey respondents who do not follow the compensatory model assumed by utility theory and conjoint analysis but, instead, follow some sort of screening or non-compensatory choice rule may have high likelihood. We may not be able to eliminate this type of respondent simply on the basis of in-sample fit.
Note that there is only a very weak correlation between time spent on the survey and the marginal log-likelihood value.
This is from a three component mixture of multivariate normals. A somewhat tighter prior was used with \(\nu =dim\left (\beta \right )+25\) and the diagonal elements of V set to .5 for all part-worths except the price partworth (transformed) that has a diagonal element of .05. This prior has much thinner tails than our default settings. Without these tighter settings, the mixture of normals model will product a large enough mass of respondents with very low price sensitivity and, even with the restricted sample, we will obtain some very large equilibrium prices (> $500).
References
Allenby, G.M., Brazell, J., Howell, J., Rossi, P.E. (2014). Valuation of Patented Product Features. Journal of Law and Economics, forthcoming.
Berry, S., Levinsohn, J., Pakes, A. (1995). Automobile prices in market equilibrium. Econometrica, 63 (4), 841–890.
Brazell, J., Diener, C., Karniouchina, E., Moore, W., Severin, V., Uldry, P.-F. (2006). The no-choice option and dual response choice designs. Marketing Letters, 17 (4), 255–268.
Chintagunta, P.K., & Nair, H. (2011). Discrete-choice models of consumer demand in marketing. Marketing Science, 30 (6), 977–996.
Choi, S.C., & DeSarbo, W.S. (1994). A conjoint-based product designing procedure incorporating price competition. Journal of Product Innovation Management, 11, 451–459.
Ding, M., Grewal, R., Liechty, J. (2005). Incentive-aligned conjoint. Journal of Marketing Research, 42 (1), 67–82.
Dong, S., Ding, M., Huber, J. (2010). A simple mechanism to incentive-align conjoint experiments. International Journal of Research in Marketing, 27 (25-32).
Horsky, D., & Nelson, P. (1992). New brand positioning and pricing in an oligolopolistic market. Marketing Science, 11 (2), 133–153.
McFadden, D.L. (1981). In M. Intrilligator, Z. Griliches, D.L. McFadden (Eds.) Econometric models of probabilistic choice (pp. 1395–1457). North-Holland.
Ofek, E., & Srinivasan, V. (2002). How much does the market value an improvement in a product attribute. Marketing Science, 21 (4), 398–411.
Orme, B.K. (2001). Assessing the monetary value of attribute levels with conjoint analysis. Discussion paper, Sawtooth Software, Inc.
Orme, B.K. (2009). Getting started with conjoint analysis. Research Publishers, LLC.
Petrin, A. (2002). Quantifying the benefits of new products: The case of the Minivan. Journal of Political Economy, 110 (4), 705–729.
Rossi, P.E. (2014). Bayesian non- and sem-parametric methods and applications, The Econometric and Tinbergen Institutes Lectures. Princeton: Princeton University Press.
Rossi, P.E., Allenby, G.M., McCulloch, R.E (2005). Bayesian statistics and marketing. Wiley.
Rossi, P.H., Wright, J.D., Anderson, A.B. (1983). Handbook of survey research. New York: Academic Press.
Sonnier, G., Ainslie, A., Otter, T. (2007). Heterogeneity distributions of willingness-to-pay in choice models. Quantitative Marketing and Economics, 5, 313–331.
Srinivasan, V., Lovejoy, W.S., Beach, D. (1997). Integrated product design for marketability and manufacturing. Journal of Marketing Research, 34 (1), 154–163.
Train, K.E. (1998). Recreational demand models with taste differences over people. Land Economics, 74 (2), 230–239.
Trajtenberg, M. (1989). The welfare analysis of product innovations, with an application to computed tomography scanners. Journal of Political Economy, 97 (2), 444–479.
Uldry, P., Severin, V., Diener, C. (2002). Using a Dual Response Framework in Choice Modeling. In AMA Advanced Research Techniques Forum.
Acknowledgments
Rossi would like to acknowledge the Collins Chair, Anderson School of Management, UCLA for research funding. Allenby thanks the Fisher College of Business at Ohio State University for generous research support. All correspondence may be addressed to the authors at the UCLA, Anderson School of Management, 110 Westwood Plaza, Los Angeles, CA 90095; or via e-mail at perossichi@gmail.com.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Allenby, G.M., Brazell, J.D., Howell, J.R. et al. Economic valuation of product features. Quant Mark Econ 12, 421–456 (2014). https://doi.org/10.1007/s11129-014-9150-x
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11129-014-9150-x