Skip to content
Licensed Unlicensed Requires Authentication Published by De Gruyter July 9, 2021

Complementarity of Information Products

Andrew T. Ching EMAIL logo , Ignatius Horstmann and Hyunwoo Lim

Abstract

In “Marketing Information: A Competitive Analysis,” Sarvary, M., and P. M. Parker. 1997. “Marketing Information: A Competitive Analysis.” Marketing Science 16 (1): 24–38 (S&P) argue that in part of the parameter space that they considered, a reduction in the price of one information product can lead to an increase in demand for another information product, i.e. information products can be gross complements. This result is surprising and has potentially important marketing implications. We show that S&P obtain this complementarity result by implicitly making the following internally inconsistent assumptions: (i) after purchasing information products, consumers update their beliefs using a Bayesian updating rule that assumes they have a diffuse initial prior (i.e. their initial prior variance is ∞ before receiving any information); (ii) if consumers choose not to purchase any information product, it is assumed that their initial prior variance is 1 (implied by the utility function specification). This internal inconsistency leads to the possibility that when information products are uncorrelated and their variances are close to 1, marginal utility is increasing in the number of products purchased, and hence information products can be complements in their model. We show that if we remove this internal inconsistency, in the parameter space considered by S&P, information products cannot be complements because the marginal utility of information products will be diminishing. We also show that, in parts of the parameter space not considered by S&P, it is possible that information products are complements; this space of parameters requires consumer’s initial prior to be relatively precise and information products to be highly correlated (either positively or negatively).

JEL Classification: D11; D81; D83; L15; L86; M31

Corresponding author: Andrew T. Ching, Carey Business School, Johns Hopkins University, Baltimore, MD, USA, E-mail:
We thank the editor, Ernan Haruvy, and four anonymous reviewers for their constructive comments and suggestions, which helped us to significantly improve this paper. We also thank Preyas Desai, Liang Guo, Susumu Imai, Xi Li, Pinar Yildirim, Juanjuan Zhang, Yi Zhu and participants at Marketing Science Conference for their helpful comments. An early version of this paper was circulated under the title “Can Information Products be Complements?”
Appendix A

We explain how we derive the intervals in the example discussed in Subsection 2.2.

Claim 1:

A consumer buys no report if and only if her type, θ ( 0 , 1 6 ) .

Proof. Note that a consumer buys no report if and only if her consumer surplus of buying no report is (i) greater than buying one report, and (ii) greater than buying both reports. We can formulate the necessary and sufficient condition for buying no report as CS (no report) > CS (both reports) & CS (no report) > CS (one report), and derive the interval for the corresponding consumer type as follows:

C S ( no report ) > C S ( both reports ) & C S ( no report ) > C S ( one report )

C S ( no report ) C S ( both reports ) > 0 & C S ( no report ) C S ( one report ) > 0

0 0.6 θ 0.1 > 0 & 0 0.2 θ 0.05 > 0  from Equations  3 a , 3 b & 3 c

0.1 0.6 θ > 0 & 0.05 0.2 θ > 0

θ < 1 6 & θ < 1 4

θ < 1 6 .

Therefore, a consumer buys no report if and only if her type (θ) is on the interval ( 0 , 1 6 ) . □

Claim 2:

A consumer buys both reports if and only if her type, θ [ 1 6 , 1 ) .

Proof. Note that a consumer buys both reports if and only if her consumer surplus of buying both reports is (i) greater than buying one report, and (ii) greater than buying no report. We can formulate the necessary and sufficient condition for buying both reports as CS (both reports) ≥ CS (one report) & CS (both report) ≥ CS (no report), and derive the interval for the corresponding consumer type as follows:

C S ( both reports ) C S ( one report ) & C S ( both reports ) C S ( no report )

C S ( both reports ) C S ( one report ) 0 & C S ( both reports ) C S ( no report ) 0

0.6 θ 0.1 0.2 θ 0.05 0   &   0.6 θ 0.1 0 0     from Eqs.  3 a , 3 b & 3 c

0.4 θ 0.05 0 & 0.6 θ 0.1 0

θ 1 8 & θ 1 6

θ 1 6 .

Therefore, a consumer buys both reports if and only if her type (θ) is on the interval [ 1 6 , 1 ) . □

Appendix B

Here, we formally show how to derive pure strategy equilibrium prices in our model. The logic of the proof is similar to S&P and Tirole (1988, p. 96–97). In the model, each consumer faces four different choices: (i) buying both products, (ii) product 1 only, (iii) product 2 only, and (iv) no product, and chooses one of them to maximize her consumer surplus. Note that we can simply regard buying both products as buying a high quality composite product. Let’s define product i’s quality, s i : = σ 0 2 σ i ˜ 2 and the composite product’s quality, S : = σ 0 2 Σ ˜ 2 . Note that S > s i for i = 1, 2. We will discuss the following five possible scenarios, where we characterize consumer demand schedules by the relationships among “quality per dollar” of products, S p 1 + p 2 , s 1 p 1 , and s 2 p 2 . Then, we will show, out of five scenarios, only scenarios 2 and 5 lead to an equilibrium. The equilibrium in scenario 2 corresponds to information products being substitutes; the equilibrium in scenario 5 corresponds to information products being complements. Without loss of generality, we assume σ 1 2 > σ 2 2 (i.e. s 2 > s 1 ) in what follows.

Scenario 1:

Suppose s 2 p 2 > s 1 p 1 S p 1 + p 2 . In this scenario, buying product 1 only is “dominated” by buying product 2 only, i.e. no consumer will buy product 1 only.[11]

Given prices p 1 and p 2, a consumer type, θ′, is indifferent between buying both products and product 2 only iff θ S p 1 p 2 = θ s 2 p 2 . A consumer type, θ″, is indifferent between buying product 2 only and no product iff θ s 2 p 2 = 0 . From these conditions, it is straightforward to show θ = p 1 S s 2 and θ = p 2 s 2 . Because S > s 2 , consumer types (θ) on the interval [ θ , 1 ) buy both products, those on the interval [ θ , θ ) buy product 2 only, and those on the interval ( 0 , θ ) buy no product.

Because the demand for the composite good is D 1,2 = ( 1 θ ) , the demand for product 1 is D 1 = D 1,2 + D 1 _ o n l y = ( 1 θ ) + 0 = 1 θ . The demand for product 2 is D 2 = D 1,2 + D 2 _ o n l y = ( 1 θ ) + ( θ θ ) = 1 θ .

The demand functions for products 1 and 2 can be re-written as:

D 1 = 1 p 1 S s 2 , D 2 = 1 p 2 s 2 .

Firm 1 maximizes the following profit: π 1 = p 1 ( 1 p 1 S s 2 ) = p 1 p 1 2 S s 2 .

Firm 2 maximizes the following profit: π 2 = p 2 ( 1 p 2 s 2 ) = p 2 p 2 2 s 2 .

The candidate equilibrium prices are:

p 1 * = S s 2 2 π 1 p 1 = 1 2 p 1 S s 2 ,

p 2 * = s 2 2    π 2 p 2 = 1 2 p 2 s 2 .

Now we want to check whether the candidate equilibrium prices satisfy the supposition. Let’s consider the first inequality of the supposition, s 2 p 2 > s 1 p 1 . By substituting p 1 * and p 2 * into it, we have

2 s 2 s 2 > 2 s 1 S s 2

2 > 2 · s 1 S s 2

S s 2 s 1 S s 2 > 0

(B1) S > s 1 + s 2 .

Let’s consider the second inequality of the supposition, s 1 p 1 S p 1 + p 2 s 1 p 1 p 1 + p 2 · S .

The supposition also states s 2 p 2 > S p 1 + p 2 s 2 > p 2 p 1 + p 2 · S .

The above two inequalities lead to s 1 + s 2 > S , which contradicts Condition (B1), S > s 1 + s 2 . This shows that these candidate equilibrium prices contradict the supposition. Therefore, this scenario does not lead to an equilibrium.

Scenario 2:

Suppose s 1 p 1 s 2 p 2 S p 1 + p 2 . In this scenario, no product is dominated.

Given prices p 1 and p 2, a consumer type, θ′, is indifferent between buying both products and product 2 only iff θ S p 1 p 2 = θ s 2 p 2 . A consumer type, θ″, is indifferent between buying product 2 only and product 1 only iff θ s 2 p 2 = θ s 1 p 1 . A consumer type, θ‴, is indifferent between buying product 1 only and no product iff θ s 1 p 1 = 0 . From these conditions, it is straightforward to show θ = p 1 S s 2 , θ = p 2 p 1 s 2 s 1 and θ = p 1 s 1 , and θ > θ ′′ > θ .

Because S > s 2 > s 1 , consumer types (θ) on the interval [ θ , 1 ) buy both products, those on the interval [ θ , θ ) buy product 2 only, those on the interval [ θ , θ ′′ ) buy product 1 only, and those on the interval 0 , θ ' ' ' buy no product.

Because the demand for the composite good is D 1,2 = ( 1 θ ) , the demand for product 2 is:

D 2 = D 1,2 + D 2 _ o n l y = ( 1 θ ) + ( θ θ ) = 1 θ .

The demand for product 1 is:

D 1 = D 1,2 + D 1 _ o n l y = 1 θ + θ ′′ θ .

Therefore, the demand for product 1 and product 2 under this scenario can be re-written as:

D 2 = 1 p 2 p 1 s 2 s 1 , D 1 = 1 p 1 S s 2 + p 2 p 1 s 2 s 1 p 1 s 1 .

Firm 2 maximizes the following profit: π 2 = p 2 1 p 2 p 1 s 2 s 1 = p 2 s 2 s 1 + p 1 s 2 s 1 p 2 2 1 s 2 s 1 .

Firm 1 maximizes the following profit:

π 1 = p 1 1 p 1 S s 2 + p 2 p 1 s 2 s 1 p 1 s 1 = p 1 s 2 s 1 + p 2 s 2 s 1 p 1 2 s 2 s 1 s 1 + S s 2 s 1 + S s 2 s 2 s 1 S s 2 s 2 s 1 s 1 .

The best response of firm 2 to p 1:

p 2 ( p 1 ) = s 2 s 1 + p 1 2 ,    π 2 p 2 = s 2 s 1 + p 1 2 p 2 s 2 s 1 .

The best response of firm 1 to p 2:

p 1 ( p 2 ) = 1 2 ( s 2 s 1 + p 2 ) ( S s 2 ) s 1 ( s 2 s 1 ) s 1 + ( S s 2 ) s 1 + ( S s 2 ) ( s 2 s 1 ) ,    π 1 p 1 = s 2 s 1 + p 2 s 2 s 1 2 p 1 ( s 2 s 1 ) s 1 + ( S s 2 ) s 1 + ( S s 2 ) ( s 2 s 1 ) ( S s 2 ) ( s 2 s 1 ) s 1 .

The candidate equilibrium prices are:

p 2 * = 3 ( S s 2 ) ( s 2 s 1 ) s 1 + 2 ( s 2 s 1 ) 2 ( S s 2 + s 1 ) 4 ( s 2 s 1 ) s 1 + 4 ( S s 2 ) s 1 + 3 ( S s 2 ) ( s 2 s 1 ) ,

p 1 * = 3 ( S s 2 ) ( s 2 s 1 ) s 1 4 ( s 2 s 1 ) s 1 + 4 ( S s 2 ) s 1 + 3 ( S s 2 ) ( s 2 s 1 ) .

Now we want to check whether the candidate prices satisfy the supposition: s 1 p 1 s 2 p 2 S p 1 + p 2 . Let’s consider the first inequality of the supposition, s 1 p 1 s 2 p 2 . By substituting p 1 * and p 2 * into it, we obtain

s 1 p 1 * s 2 p 2 *

s 1 p 2 * s 2 p 1 *

s 1 p 2 * s 2 p 1 * 0

3 ( s 1 s 2 ) ( S s 2 ) ( s 2 s 1 ) s 1 + 2 s 1 ( s 2 s 1 ) 2 ( S s 2 + s 1 ) 0

( s 2 s 1 ) 2 ( 3 ( S s 2 ) s 1 + 2 s 1 ( S s 2 + s 1 ) ) 0

s 1 ( s 2 s 1 ) 2 ( 3 S + 3 s 2 + 2 S 2 s 2 + 2 s 1 ) 0

s 1 ( s 2 s 1 ) 2 ( 2 s 1 + s 2 S ) 0

(B2) 2 s 1 + s 2 S

Let’s consider the second inequality of the supposition, s 2 p 2 S p 1 + p 2 . By substituting p 1 * and p 2 * into it, we obtain

s 2 p 2 * S p 1 * + p 2 *

s 2 ( p 1 * + p 2 * ) S p 2 * 0

s 2 p 1 * + ( s 2 S ) p 2 * 0

s 2 3 S s 2 s 2 s 1 s 1 + s 2 S ( 3 S s 2 s 2 s 1 s 1 + 2 s 2 s 1 2 S s 2 + s 1 ) 0

3 2 s 2 S S s 2 s 2 s 1 s 1 + 2 s 2 S s 2 s 1 2 S s 2 + s 1 0

( s 2 s 1 ) ( S s 2 ) ( 3 s 1 ( 2 s 2 S ) 2 ( s 2 s 1 ) ( S s 2 + s 1 ) ) 0

s 2 s 1 S s 2 3 s 1 2 s 2 + 2 s 1 S + 6 s 1 s 2 + 2 s 2 s 1 2 0

2 s 1 2 + 2 s 2 2 + 2 s 1 s 2 s 1 + 2 s 2 S

(B3) 2 s 1 + s 2 3 s 1 s 2 s 1 + 2 s 2 S

It is clear that condition (B3) is stronger than condition (B2). We show that the candidate equilibrium derived above is indeed an equilibrium iff s 2 p 2 * S p 1 * + p 2 * is satisfied.

We can also confirm that information products are substitutes in this scenario from D 2 p 1 = 1 s 2 s 1 > 0 and D 1 p 2 = 1 s 2 s 1 > 0 .

Scenario 3:

Suppose s 1 p 1 > S p 1 + p 2 s 2 p 2 . In this scenario, buying product 2 only is “dominated” by the composite product, i.e. no consumer will buy product 2 only.

Given prices p 1 and p 2, a consumer type, θ′, is indifferent between buying both products and product 1 only iff θ S p 1 p 2 = θ s 1 p 1 . A consumer type, θ″, is indifferent between buying product 1 only and no product iff θ s 1 p 1 = 0 . From the conditions, it is straightforward to show θ = p 2 S s 1 and θ = p 1 s 1 .

Because S > s 1 , consumer types (θ) on the interval [ θ , 1 ) buy both products, those on the interval [ θ , θ ) buy product 1 only, and those on the interval ( 0 , θ ) buy no product. Because the demand for the composite good is D 1,2 = ( 1 θ ) , the demand for product 1 is D 1 = D 1,2 + D 1 _ o n l y = ( 1 θ ) + ( θ θ ) = 1 θ . The demand for product 2 is D 2 = D 1,2 + D 2 _ o n l y = ( 1 θ ) + 0 = ( 1 θ ) .

The demand functions for products 1 and 2 can be re-written as:

D 1 = 1 p 1 s 1 , D 2 = 1 p 2 S s 1 .

Firm 1 maximizes the following profit:

π 1 = p 1 ( 1 p 1 s 1 ) = p 1 p 1 2 s 1 .

Firm 2 maximizes the following profit:

π 2 = p 2 ( 1 p 2 S s 1 ) = p 2 p 2 2 S s 1 .

The candidate equilibrium prices are:

p 1 * = s 1 2    π 1 p 1 = 1 2 p 1 s 1 ,

p 2 * = S s 1 2    π 2 p 2 = 1 2 p 2 S s 1 .

Now we want to check whether the candidate prices satisfy the supposition. Let’s consider the first inequality of the supposition, s 1 p 1 > S p 1 + p 2 . By substituting p 1 * and p 2 * into it, we obtain s 1 p 1 * > S p 1 * + p 2 * s 1 s 1 2 > S s 1 + S s 1 2 2 > 2 . This shows that these candidate equilibrium prices contradict the supposition. Therefore, this scenario does not lead to an equilibrium.

Scenario 4:

Suppose s 2 p 2 > S p 1 + p 2 s 1 p 1 . A similar argument as Scenario 3 holds for this scenario. Therefore, this scenario does not lead to an equilibrium.

Scenario 5:

Suppose S p 1 + p 2 > s 1 p 1 and S p 1 + p 2 > s 2 p 2 . In this scenario, both buying product 2 only and buying product 1 only are “dominated” by the composite product, i.e. no consumer will buy product 1 only or product 2 only.

Given prices p 1 and p 2, a consumer type, θ′, is indifferent between buying both products and no product iff θ S p 1 p 2 = 0 . From the condition, it is straightforward to show θ = p 1 + p 2 S .

Consumer types (θ) on the interval [ θ , 1 ) buy both products, those on the interval [ 0 , θ ) buy no product.

Because the demand for the composite good is D 1,2 = ( 1 θ ) , the demand for product 1 is D 1 = D 1,2 + D 1 _ o n l y = ( 1 θ ) + 0 = 1 θ . The demand for product 2 is D 2 = D 1,2 + D 2 _ o n l y = ( 1 θ ) + 0 = 1 θ .

The demand functions for products 1 and 2 can be re-written as:

D 1 = 1 p 1 + p 2 S , D 2 = 1 p 1 + p 2 S .

Firm 1 maximizes the following profit: π 1 = p 1 1 p 1 + p 2 S = p 1 · S p 1 p 2 S .

Firm 2 maximizes the following profit: π 2 = p 2 1 p 1 + p 2 S = p 2 · S p 1 p 2 S .

The best response of firm 1 to p 2:

p 1 ( p 2 ) = S p 2 2 ,    π 1 p 1 = 2 p 1 + S p 2 S .

The best response of firm 2 to p 1:

p 2 ( p 1 ) = S p 1 2 ,    π 2 p 2 = 2 p 2 + S p 1 S .

The candidate equilibrium prices are:

p 1 * = p 2 * = S 3 .

Now we want to check whether the candidate prices satisfy the supposition. Let’s consider the first inequality of the supposition, S p 1 + p 2 > s 1 p 1 . By substituting p 1 * and p 2 * into it, we obtain

S p 1 * + p 2 * > s 1 p 1 *

S 2 S 3 > s 1 S 3

3 2 > 3 s 1 S

S > 2 s 1 .

By substituting p 1 * and p 2 * into the second inequality of the supposition, S p 1 + p 2 > s 2 p 2 , we obtain

S p 1 * + p 2 * > s 2 p 2 *

S 2 S 3 > s 2 S 3

3 2 > 3 s 2 S

S > 2 s 2 .

Hence, we show that this scenario will lead to an equilibrium iff S > 2 s 1 and S > 2 s 2 hold. We also confirm that information products are complements in this scenario from D 2 p 1 = 1 S < 0 and D 1 p 2 = 1 S < 0 .

References

Admati, A. R., and P. Pfleiderer. 1987. “Viable Allocations of Information in Financial Markets.” Journal of Economic Theory 43 (1): 76–115, https://doi.org/10.1016/0022-0531(87)90116-5.Search in Google Scholar

Arora, A., and A. Fosfuri. 2005. “Pricing Diagnostic Information.” Management Science 51 (7): 1092–100, doi:https://doi.org/10.1287/mnsc.1050.0362.Search in Google Scholar

Banerjee, S., J. Davis, and N. Gondhi. 2018. “When Transparency Improves, Must Prices Reflect Fundamentals Better?” The Review of Financial Studies 31 (6): 2377–414, https://doi.org/10.1093/rfs/hhy034.Search in Google Scholar

Bergemann, D., and A. Bonatti. 2015. “Selling Cookies.” American Economic Journal: Microeconomics 7 (3): 259–94, https://doi.org/10.1257/mic.20140155.Search in Google Scholar

Chen, Y., C. Narasimhan, and Z. J. Zhang. 2001. “Individual Marketing with Imperfect Targetability.” Marketing Science 20 (1): 23–41, doi:https://doi.org/10.1287/mksc.20.1.23.10201.Search in Google Scholar

Ching, A. T., and H. Lim. 2020. “A Structural Model of Correlated Learning and Late-Mover Advantages: The Case of Statins.” Management Science 66 (3): 1095–123, https://doi.org/10.1287/mnsc.2018.3221.Search in Google Scholar

Ching, A. T., R. Clark, I. Horstmann, and H. Lim. 2016. “The Effects of Publicity on Demand: The Case of Anti-Cholesterol Drugs.” Marketing Science 35 (1): 158–81.10.1287/mksc.2015.0925Search in Google Scholar

Christen, M. 2005. “Research Note-Cost Uncertainty is Bliss: The Effect of Competition on the Acquisition of Cost Information for Pricing New Products.” Management Science 51 (4): 668–76, doi:https://doi.org/10.1287/mnsc.1040.0320.Search in Google Scholar

Christen, M., and M. Sarvary. 2007. “Competitive Pricing of Information: A Longitudinal Experiment.” Journal of Marketing Research 44 (1): 42–56, https://doi.org/10.1509/jmkr.44.1.42.Search in Google Scholar

Gal-Or, E., T. Geylani, and T. P. Yildirim. 2012. “The Impact of Advertising on Media Bias.” Journal of Marketing Research 49 (1): 92–9, doi:https://doi.org/10.1509/jmr.10.0196.Search in Google Scholar

Goldstein, I., and L. Yang. 2015. “Information Diversity and Complementarities in Trading and Information Acquisition.” The Journal of Finance 70 (4): 1723–65, https://doi.org/10.1111/jofi.12226.Search in Google Scholar

Guo, L. 2006. “Consumption Flexibility, Product Configuration, and Market Competition.” Marketing Science 25 (2): 116–30, https://doi.org/10.1287/mksc.1050.0169.Search in Google Scholar

Guo, L., and Y. Zhao. 2009. “Voluntary Quality Disclosure and Market Interaction.” Marketing Science 28 (3): 488–501, https://doi.org/10.1287/mksc.1080.0418.Search in Google Scholar

Iyer, G., and D. Soberman. 2000. “Markets for Product Modification Information.” Marketing Science 19 (3): 203–25, https://doi.org/10.1287/mksc.19.3.203.11801.Search in Google Scholar

Jensen, F. O. 1991. “Information Services.” In The AMA Handbook of Marketing for the Service Industries, chap. 22, edited by M. L. Congram, and C. A. Friedman, 423–43. New York: American Management Association.Search in Google Scholar

Ke, T. T., and S. Lin. 2020. “Informational Complementarity.” Management Science 66 (8): 3699–716, https://doi.org/10.1287/mnsc.2019.3377.Search in Google Scholar

Lambrecht, A., A. Goldfarb, A. Bonatti, A. Ghose, D. G. Goldstein, R. Lewis, A. Rao, N. Sahni, and S. Yao. 2014. “How Do Firms Make Money Selling Digital Goods Online?” Marketing Letters 25 (3): 331–41, https://doi.org/10.1007/s11002-014-9310-5.Search in Google Scholar

Murphy, K. P. 2007. Conjugate Bayesian Analysis of the Gaussian Distribution. Tech. Rep. Also available at http://www.cs.ubc.ca/∼murphyk/Papers/bayesGauss.pdf.Search in Google Scholar

Raju, J. S., and A. Roy. 2000. “Market Information and Firm Performance.” Management Science 46 (8): 1075–84, https://doi.org/10.1287/mnsc.46.8.1075.12024.Search in Google Scholar

Sarvary, M. 2002. “Temporal Differentiation and the Market for Second Opinions.” Journal of Marketing Research 39 (1): 129–36, https://doi.org/10.1509/jmkr.39.1.129.18933.Search in Google Scholar

Sarvary, M., and P. M. Parker. 1997. “Marketing Information: A Competitive Analysis.” Marketing Science 16 (1): 24–38, https://doi.org/10.1287/mksc.16.1.24.Search in Google Scholar

Tirole, J. 1988. The Theory of Industrial Organization. Cambridge: The MIT Press.Search in Google Scholar

Winkler, R. L. 1981. “Combining Probability Distributions from Dependent Information Sources.” Management Science 27 (4): 479–88, https://doi.org/10.1287/mnsc.27.4.479.Search in Google Scholar

Xiang, Y., and M. Sarvary. 2013. “Buying and Selling Information Under Competition.” Quantitative Marketing and Economics 11 (3): 321–51, https://doi.org/10.1007/s11129-013-9135-1.Search in Google Scholar

Received: 2021-01-21
Accepted: 2021-06-03
Published Online: 2021-07-09

© 2021 Walter de Gruyter GmbH, Berlin/Boston

Downloaded on 8.12.2022 from https://www.degruyter.com/document/doi/10.1515/roms-2021-0004/html
Scroll Up Arrow