Home The Role of Informative Advertising in Aligning Preferences Over Product Design
Article
Licensed
Unlicensed Requires Authentication

The Role of Informative Advertising in Aligning Preferences Over Product Design

  • Guillem Roig EMAIL logo
Published/Copyright: June 24, 2024

Abstract

This article explores the relationship between product design and advertising, traditionally viewed as distinct business decisions. I build upon the seminal work of Matutes and Regibeau (1988. “Mix and Match: Product Compatibility Without Network Externalities.” The RAND Journal of Economics 19 (2): 221–34), where firms produce components that combine to form product systems, and introduce the necessity for firms to advertise their products to create consumer awareness. I find that a firm’s incentive to advertise increases with product compatibility. Greater product compatibility reduces competition, encouraging firms to broaden their market presence through advertising. Contrary to Matutes and Regibeau, when advertising influences product awareness, consumers prefer products with moderate compatibility over those that are entirely incompatible. Consequently, advertising aligns the compatibility preferences of both firms and consumers.

JEL Classification: D21; D43; L13; L15

Corresponding author: Guillem Roig, Faculty of Economics, Universidad Del Rosario, Bogotá, Colombia, E-mail: 

Appendix

Proof of Proposition (1)

We proceed to show that firms have no profitable deviation form the advertising reach and prices established in the proposition.

  • Partially-informed market, ϕ* < 1:

    The equilibrium price is p * = 2 / α ( 1 + z ) . Setting any price other than p* in the vicinity is unprofitable, as it will not be in the best response. Setting a price p′ = (V − 2)/2 such that only captive consumers purchase is not profitable whenever:

    ϕ * p * + ϕ * ( 1 ϕ * ) 2 p * > 2 ϕ * ( 1 ϕ * ) ( V 2 ) / 2 .

    Rearranging terms, this is equivalent to p*(2 − ϕ*)/(1 − ϕ*) + 2 > V, and substituting for the equilibrium price and advertising reach gives V < ( 4 + 2 2 α ( 1 + z ) 2 α ) / ( 2 α ( 1 + z ) α ) V ̄ .

    A fully-covered market implies that V ≥ 2p* + 2. Substituting for the equilibrium prices gives V ( 2 2 + 2 α ( 1 + z ) ) / ( α ( 1 + z ) ) V ̲ .

    It is easy to show that for any α > 0, we get V ̄ > V ̲ since

    ( 4 + 2 2 α ( 1 + z ) 2 α ) / ( 2 α ( 1 + z ) α ) > ( 2 2 + 2 α ( 1 + z ) ) / ( α ( 1 + z ) ) 4 + 2 2 η 2 α ) / ( 2 η α ) > ( 2 2 + 2 η ) / η 0 > 2 2 α ,

    where the second lines comes from setting η α ( 1 + z ) . Hence, there is no profitable deviation in prices as long as V < V ̄ .

    A deviation in adverting in the neighborhood of ϕ*, will not be profitable as it will not belong to the firm’s best response, and hence, we consider the extremes. A deviation consisting in setting ϕ′ = 0 is clearly not profitable as no consumer will be aware of the product. Setting ϕ′ = 1 is not profitable whenever:

    p * ( ϕ * ) 2 + 2 ϕ * ( 1 ϕ * ) ϕ * 2 ( 1 ϕ * ) E ( ϕ * ) E ( ϕ = 1 ) 1 α ( ϕ * ) 2 α p * 2 + ( ϕ * ) 2 3 ϕ * 1 2 α α + 2 η α 2 η 2 + 4 α 2 ( α + 2 η ) 2 6 α α + 2 η ,

    where the second line comes from introducing the equilibrium advertising reach and setting η α ( 1 + z ) . The previous inequality reduces to 4 η 2 + 3 α 2 4 2 α η . Then, since η α ( 1 + z ) and arranging terms gives 4 ( 1 + z ) + 3 α 2 4 2 2 α ( 1 + z ) , which is equivalent to 16(1 + z)2 + 24(1 + z)α 2 + 9α 4 ≥ 32α(1 + z), and this is fulfilled for any z and α.

  • Fully-informed market, ϕ* = 1:

    The equilibrium price is p* = 1/(1 + z). Setting any price other than p* in the vicinity is unprofitable, as it will not be in the best response. Moreover, pricing at p′ = (V − 2)/2 is also unprofitable due to the absence of captive consumers. The market is fully-covered whenever V ≥ 1 + 2p* which gives V ≥ (3 + z)/(1 + z), and this is fulfilled for any V ≥ 3.

    A deviation in advertising reach, ϕ′ = 1 − ϵ for ϵ ∈ (0, 1], results in profit loss of 1/2(1 − ϵ)2p* − (1/2)2p* = −ϵp*, while gains from lower advertising costs are ϵ 2/α. A deviation is unprofitable if ϵp* > ϵ 2/α which is equivalent to α > ϵ(1 + z), which is always true for ϵ ∈ (0, 1] given that in a fully-informed market α > 2(1 + z).

Proof of Proposition (2).

To establish that an intermediate level of compatibility maximizes consumer surplus, I differentiate CS(⋅) with respect to z to show that CS(⋅) is a quasi-concave function in z. Then:

d C S ( z , α ) d z = 2 α 2 8 2 + α ( 1 + z ) × z α ( 2 3 z ) α 10 + 2 α × 3 2 z z + z 2 1 α ( 1 + z ) × α + 2 α ( 1 + z ) 3 .

Evaluating the function at both extremes gives d C S ( z = 0 , α ) d z = 2 α 8 2 + 3 2 α α ( 10 + α ) ( 2 + α ) 3 , and d C S ( z = 1 , α ) d z = 2 α 8 2 + 2 α 2 α ( 10 + 2 α ) ( 2 + α ) 3 . The derivative evaluated at z = 0 is positive if 8 2 + 3 2 α α ( 10 + α ) > 0 The derivative evaluated at z = 1 is negative if 8 2 + 2 α 2 α ( 10 + 2 α ) < 0 . Define α′ as the solution of 8 2 = α ( 10 + α ) 3 2 α , and α″ as the solution of 8 = α ( 10 + 2 α ) α . Since 2 α ( 10 + 2 α ) 2 α > α ( 10 + α ) 3 2 α , it implies that α″ < α′, and there will be a maximum at an intermediate level of product compatibility for α ∈ (α″, α′). Then, it is only left to show that this range for α is included in the case of a partially-informed market, or equivalently, α < α ̄ ( z ) .

We first show that α > α ̄ ( z ) . To see this, use 8 2 = α ( 10 + α ) 3 2 α and define RHS ( α ) α ( 10 + α ) 3 2 α . Since RHS ( α ) / α = 5 / α + 3 / 2 α 3 / 2 > 0 , then, replacing for the highest possible α which is α ̄ ( z = 1 ) = 4 , we get 8 2 > 4 ( 10 + 4 ) 3 2 × 4 . This implies that the solution α′ must be larger than α ̄ ( z ) .

To show that α < α ̄ ( z ) use 8 = α ( 10 + 2 α ) α . Define RHS ( α ) α ( 10 + 2 α ) α . Since RHS ( α ) / α = 10 / α + 3 α 1 > 0 , then, replacing for the highest possible α which is α ̄ ( z = 1 ) = 4 , we get 8 < 4 ( 10 + 8 ) 4 . Then the solution α″ must be smaller than α ̄ ( z ) .

This demonstrates that there exists an ϵ > 0 such that for α α , α + ϵ , consumer welfare is maximized at an intermediate level of product compatibility as stated in the proposition.

Preference Costs

Captive consumers know of the existence of only one system and their average preference cost is

P C c = 0 1 0 1 ( x + y ) d y d x = 0 1 0 1 x y + y 2 2 d x = 0 1 x 2 2 + x 2 = 1 .

Selective consumers are aware of both systems and can mix and match components from different firms. The average preference cost is determined by assigning selective consumers to distinct exclusive locations, as depicted in the figure. Parentheses indicate consumers’ consumption

Computing the areas from the different regions results in

P C s ( z ) = 2 ( a + b + c + d ) = 2 0 1 z 2 0 1 z 2 ( x + y ) d y d x + 0 1 z 2 0 1 + z 2 ( x + y ) d y d x + 2 1 z 2 1 + z 2 0 1 z 2 ( x + y ) d y + 1 z 2 1 x ( x + y ) d y d x = 2 ( 1 z ) 3 8 a + 1 z 2 8 b + z ( 1 z ) ( 3 z ) 8 c + z 2 ( 3 z ) 6 d = 3 + z 2 ( 3 2 z ) 6 .

References

Bagwell, K. 2007. “The Economic Analysis of Advertising.” In Handbook of Industrial Organization, Vol. 3. Chapter 28. North-Holland: Elsevier.10.1016/S1573-448X(06)03028-7Search in Google Scholar

Butters, G. 1977. “Equilibrium Distribution of Sales and Advertising Prices.” The Review of Economic Studies 44 (3): 465–91. https://doi.org/10.2307/2296902.Search in Google Scholar

Chen, Z., and P. Rey. 2012. “Loss Leading as an Exploitative Practice.” The American Economic Review 102 (7): 62–82. https://doi.org/10.1257/aer.102.7.3462.Search in Google Scholar

Economides, N. 1989. “Desirability of Compatibility in the Absence of Network Externalities.” The American Economic Review 79: 1165–81.Search in Google Scholar

Ellison, G. 2005. “A Model of Add-On Pricing.” Quarterly Journal of Economics 120 (2): 585–637. https://doi.org/10.1093/qje/120.2.585.Search in Google Scholar

Grossman, G., and C. Shapiro. 1984. “Informative Advertising with Differentiated Products.” The Review of Economic Studies 51: 63–81. https://doi.org/10.2307/2297705.Search in Google Scholar

Hamilton, S. F. 2009. “Informative Advertising in Differentiated Oligopoly Markets.” International Journal of Industrial Organization 27 (1): 60–9. https://doi.org/10.1016/j.ijindorg.2008.04.002.Search in Google Scholar

Kim, S.-H., and J. P. Choi. 2015. “Optimal Compatibility in System Markets.” Games and Economic Behavior 90: 106–18. https://doi.org/10.1016/j.geb.2015.02.005.Search in Google Scholar

Matutes, C., and P. Regibeau. 1988. “Mix and Match: Product Compatibility without Network Externalities.” The RAND Journal of Economics 19 (2): 221–34. https://doi.org/10.2307/2555701.Search in Google Scholar

Roig, G. 2020. “Product Compatibility Hinders Pre-Emptive Advertising.” Economic Inquiry 58 (4): 1663–88. https://doi.org/10.1111/ecin.12887.Search in Google Scholar

Soberman, D. A. 2004. “Research Note: Additional Learning and Implications on the Role of Informative Advertising.” Management Science 50 (12): 1744–50. https://doi.org/10.1287/mnsc.1040.0288.Search in Google Scholar

Stegeman, M. 1991. “Advertising in Competitive Markets.” The American Economic Review 81 (1): 210–23.Search in Google Scholar

Received: 2023-09-07
Accepted: 2024-05-01
Published Online: 2024-06-24

© 2024 Walter de Gruyter GmbH, Berlin/Boston

Downloaded on 10.9.2025 from https://www.degruyterbrill.com/document/doi/10.1515/bejte-2023-0096/html
Scroll to top button