Skip to main content
Log in

On the Impact of Input Prices on an Entrant’s Profit Under Multi-Product Competition

  • Published:
Review of Industrial Organization Aims and scope Submit manuscript

Abstract

We study the impact of input prices on an entrant’s profit when firms are engaged in multi-product competition. We consider a setting with both horizontal and vertical differentiation, in which a vertically integrated firm controls the input that is required for the supply of the high-quality product. We establish the conditions under which the entrant is better off from an increase in the wholesale price of this critical input. This possibility contrasts with the existing literature that, under linear demands, finds a non-positive relationship between the input price and the single-product entrant’s profit.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2

Similar content being viewed by others

Notes

  1. This effect exists even if the incumbent is not a vertically integrated firm. In that case, the effect is weaker because the incumbent only raises its retail price in response to the entrant’s price increase. With a vertically integrated incumbent, however, an increase in the input price also corresponds to an increase in the incumbent’s (opportunity) cost. To the extent that an increase in its retail customers corresponds to a decrease in the entrant’s customers, selling one more unit at the retail level means forgoing some wholesale revenues.

  2. Our results would not qualitatively change if the entrant did additionally have constant marginal costs for each product.

  3. The assumption that the “second preferred” product of any consumer is never completely different from her “first preferred” one does not affect the results qualitatively, but it is made for expositional purposes.

  4. Since consumers’ decisions only depend on the perceived quality difference, one can normalize \(\delta_{iL} = 0\) without loss of generality.

  5. Our aim is to keep the model as simple as possible in order to derive quite general results. In particular, the level of the input price is considered as an exogenous factor. In Sect. 6.1 we discuss the case in which the incumbent sets the input price, and in Sect. 7 we discuss the policy implications of our results when it is set by a regulator.

  6. Indeed, \(\frac{{\partial p_{IH} }}{\partial w} = \frac{8 - 5\rho }{(4 - \rho )(4 - 3\rho )}\) and \(\frac{{\partial p_{IL} }}{\partial w} = \frac{(8 - 3\rho )(1 - \rho )}{(4 - \rho )(4 - 3\rho )}\). Hence, \(\frac{{\partial p_{IH} }}{\partial w} - \frac{{\partial p_{IL} }}{\partial w} = \frac{3\rho (2 - \rho )}{(4 - \rho )(4 - 3\rho )} > 0\).

  7. Section 6.2 shows that this finding is common in any (symmetric) linear demand model in which the demand for each product does not change if all prices increase by the same amount.

  8. Throughout the paper, we assume that the entrant buys only from the incumbent the input that is required for the production of the high-quality product. Equations (11) and (12) do not depend on this assumption. If the input was required to provide the low-quality product instead, the model could accommodate this possibility by simply assuming a negative value of \(\delta\) and appropriately rewriting Assumption 1(ii) and (iii). In this case, an increase in \(w\) benefits the entrant if the profit margin on the high-quality product exceeds the margin on the low-quality product; this is a condition that always holds because the low-quality product also has higher marginal costs.

  9. See Eq. (14) in the next section.

  10. It can be easily shown that this privately optimal level belongs to \([\underline{w} ,\bar{w}]\) for a range of combinations of \(\delta\) and \(\rho\). The same would not hold, however, if the entrant bought the critical input required for the supply of the low-quality product instead.

  11. With the entrant's prices kept constant, a higher \(w\) increases the incumbent's wholesale revenues. However, increasing \(w\) also leads to higher equilibrium prices that are set by the entrant. An increase in \(p_{EL}\) directs some consumers away from (EL) and towards (IL) and (EH). As this positively affects the incumbent's retail and wholesale revenues, an increase in \(p_{EL}\) is positive for the incumbent. An increase in \(p_{EH}\), however, may have the opposite effect. It makes consumers switch from (EH) to (IH), but it also directs consumers to (EL). Despite the incumbent's retail gains, there is a wholesale loss that dominates all of the other effects if \(w\) is sufficiently large, provided that \(\rho \in (0,1)\).

  12. A sketch of the proof is given in Appendix 3.

  13. It is very difficult to verify whether this condition holds in practice. In our motivating example, such verification requires refined data about the retail prices of the copper-based and the fiber-based services. The limited penetration of the high-quality (ultra-fast) broadband services (1.2 % at EU level (European Commission 2013)), as well as the several retail offers that are available by the firms (see, for instance, Van Dijk Management Consultants 2013) does not allow for such refined data. When the entrant has not deployed a copper access network and thus it also pays a copper access price to the incumbent, its profit margin in the copper market is less likely to be higher than its profit margin in the fiber market. However, by considering the case of the UK, where there are available data (OECD 2013), we conclude that, even in this worst case scenario, this condition has a reasonable chance of happening.

  14. In this case, banning access is the only way to lead the entrant to make the upstream input since any positive profit is preferred to the zero profit that stems from exiting the market.

  15. The reasoning of the “business migration effect” that is discussed by Bourreau et al. (2012b) partially explains this outcome.

References

  • Ailawadi, K., & Harlam, B. (2004). An empirical analysis of the determinants of retail margins: The role of store brand share. Journal of Marketing, 68(1), 147–165.

    Article  Google Scholar 

  • Armstrong, M., Doyle, C., & Vickers, J. (1996). The access pricing problem: A synthesis. The Journal of Industrial Economics, 44(2), 131–150.

    Article  Google Scholar 

  • Bourreau, M., Cambini, C., & Doğan, P. (2012a). Access pricing, competition, and incentives to migrate from “old” to “new” technology. International Journal of Industrial Organization, 30(6), 713–723.

    Article  Google Scholar 

  • Bourreau, M., Cambini, C., & Hoernig, S. (2012b). Ex ante regulation and co-investment in the transition to next generation access. Telecommunications Policy, 36(5), 399–406.

    Article  Google Scholar 

  • Chen, Y., & Riordan, M. (2007). Price and variety in the spokes model. The Economic Journal, 117(522), 897–921.

    Article  Google Scholar 

  • Deneckere, R., & McAfee, R. P. (1996). Damaged goods. Journal of Economics and Management Strategy, 5(2), 149–174.

    Article  Google Scholar 

  • Dixit, A. (1986). Comparative statics for oligopoly. International Economic Review, 27(1), 107–122.

    Article  Google Scholar 

  • European Commission. (2013). Broadband access in the EU: situation at 1 July 2013 (COCOM 14-03), Brussels.

  • Gayle, P. G., & Weisman, D. L. (2007). Are input prices irrelevant for make-or-buy decisions? Journal of Regulatory Economics, 32(2), 195–207.

    Article  Google Scholar 

  • Hoch, S. J. (1996). How should national brands think about private labels? Sloan Management Review, 37(2), 89–102.

    Google Scholar 

  • Mandy, D. M. (2008). Pricing inputs to induce efficient make-or-buy decisions. Journal of Regulatory Economics, 36(1), 29–43.

    Article  Google Scholar 

  • OECD. (2013). OECD communications outlook 2013. Paris: OECD Publishing.

    Google Scholar 

  • Rogers, E. M. (1983). Diffusion of innovation (3rd ed.). New York: The free press.

    Google Scholar 

  • Sappington, D. E. M. (2005). On the irrelevance of input prices for make-or-buy decisions. American Economic Review, 95(5), 1631–1638.

    Article  Google Scholar 

  • Van Dijk Management Consultants. (2013). Broadband Internet Access Cost (BIAC): Prices as at 1–15 February 2013. A study prepared for the European Commission DG Communications Networks, Content & Technology. Luxembourg: Publications Office of the European Union. http://ec.europa.eu/newsroom/dae/document.cfm?action=display&doc_id=4594. Retrieved 18 July 2016.

  • Wang, X. H., & Zhao, J. (2010). Why are firms sometimes unwilling to reduce costs? Journal of Economics, 101(2), 103–124.

    Article  Google Scholar 

Download references

Acknowledgments

The authors are pleased to acknowledge financial support from Fundação para a Ciência e a Tecnologia (Grant UID/ECO/04007/2013) and FEDER/COMPETE (POCI-01-0145-FEDER-007659). Tselekounis acknowledges financial support from Fundação para a Ciência e a Tecnologia (Grant SFRH/BPD/98501/2013). We also thank the editor and anonymous referees for their very useful comments and suggestions that considerably improved the paper.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Markos Tselekounis.

Appendices

Appendix 1

Throughout the paper we assume that the consumer who is indifferent between choosing products \(\left( {ij} \right)\) and \(\left( {ml} \right)\), with either \(i \ne m\) and \(j = l\) or \(i = m\) and \(j \ne l\), where \(i,m = I,E\) and \(j,l = L,H\) is located in [0,1], i.e., \(0 \le \hat{x}_{ij/ml} \le 1\). In particular:

$$0 \le \hat{x}_{IH/EH} \le 1 \Rightarrow R1 \equiv - \frac{(4 - \rho )}{(1 - \rho )} \le w \le \frac{(4 - \rho )}{(1 - \rho )} \equiv R2$$
(18)
$$0 \le \hat{x}_{IL/EL} \le 1 \Rightarrow R1 \le w \le R2$$
(19)
$$\begin{aligned} 0 &\le \hat{x}_{IH/IL} \le 1 \Rightarrow \hfill \\ R3 &\equiv \frac{{(4 - \rho )\left[ {\delta (2 - \rho ) - (4 - 3\rho )} \right]}}{3\rho (2 - \rho )} \le w \le \frac{{(4 - \rho )\left[ {\delta (2 - \rho ) + (4 - 3\rho )} \right]}}{3\rho (2 - \rho )} \equiv R4 \hfill \\ \end{aligned}$$
(20)
$$\begin{aligned} 0 & \le \hat{x}_{EH/EL} \le 1 \Rightarrow \\ R5 & \equiv \frac{{(4 - \rho )\left[ {\delta (2 - \rho ) - (4 - 3\rho )} \right]}}{8 + (3\rho - 8)\rho } \le w \le \frac{{(4 - \rho )\left[ {\delta (2 - \rho ) + (4 - 3\rho )} \right]}}{8 + (3\rho - 8)\rho } \equiv R6. \\ \end{aligned}$$
(21)

Therefore, the value of \(w\) should be such that all inequalities (18)–(21) hold, or \(max\{ R1,R3,R5,0\} \le w \le min\{ R2,R4,R6\}\). It is straightforward to show that R5 > R1 and R6 < R4. Moreover, R5 > R3 if and only if \(\delta < \underline{\delta } = (4 - 3\rho )/(2 - \rho )\) and R6 < R2 if and only if \(\delta < \overline{\delta } = (4 - \rho )/\left[ {(2 - \rho )(1 - \rho )} \right]\), with \(\overline{\delta } > \underline{\delta }\) for all \(\rho \in (0,1]\).

Hence, the condition that ensures that (18)–(21) hold is \(max\{ R3,R5\} \le w \le min\{ R2,R6\}\) and can be subdivided into three cases:

  1. 1.

    When \(\delta > \overline{\delta }\), the necessary and sufficient condition is \(R3 \le w \le R2\).

  2. 2.

    When \(\underline{\delta } < \delta < \overline{\delta }\), the necessary and sufficient condition is \(R3 \le w \le R6\).

  3. 3.

    When \(\delta < \underline{\delta }\), the necessary and sufficient condition is \(R5 \le w \le R6\).

As \(R2 > R3 \Leftrightarrow \delta < \overline{\delta }\), \(R6 > R3 \Leftrightarrow \delta < \overline{\delta }\) and \(R6 > R5\) for all parameter values, this implies that: (1) The admissible values of \(\delta\) are in the interval [\(0,\overline{\delta }\)]; and (2) The equilibrium retail prices presented in Lemma 1, as well as the derived demand function of each product hold when \(w \in [\underline{w} ,\bar{w}]\), where \(\underline{w} = max\{ R3,R5\}\) and \(\bar{w} = R6\).

We now show that the condition for the entrant’s profit to increase in the fiber access price is not inconsistent with both the upper and lower limits in Assumption 1(ii). Indeed, with respect to the upper limit, it is shown that \(\hat{w} < \bar{w}\) since

$$\bar{w} - \hat{w} = \frac{(4 - \rho )(4 - 3\rho )(4 + \delta \rho )}{{4\left[ {8 + (3\rho - 8)\rho } \right]}} > 0.$$
(22)

As for the lower limit, when \(\delta < \underline{\delta }\), \(\hat{w}\) is higher than R5 if and only if \(\delta < 4/\rho\), which always holds when \(\delta < \underline{\delta }\). When \(\delta > \underline{\delta }\), \(\hat{w}\) is higher than R3 if and only if \(\delta < \widetilde{\delta } = 4/(2 - \rho )\). Note that \(\underline{\delta } < \widetilde{\delta } < \overline{\delta }\) for \(\rho \ne 0\). Therefore, if \(\delta \in [0,\widetilde{\delta }]\), then \(\underline{w} \le \hat{w} < \bar{w}\). If \(\delta \in [\widetilde{\delta },\overline{\delta } ]\), then \(\hat{w} < \underline{w}\) which means that the entrant’s profit is positively correlated with the fiber access price, for all admissible \(w\). These results prove Proposition 1.

Appendix 2

The functions \(f_{1}\), \(f_{2}\), \(f_{3}\), \(f_{4}\), and \(f_{5}\) are as given below:

$$\begin{aligned} f_{1} & = \frac{{(1 - \rho )(2 - \rho )(9\rho^{2} - 32\rho + 32)}}{{(4 - \rho )^{2} (4 - 3\rho )^{2} }},\quad f_{2} = \frac{{(1 - \rho )(7\rho^{2} - 20\rho + 16)}}{{(4 - \rho )(4 - 3\rho )^{2} }} \\ f_{3} & = \frac{{(2 - \rho )(1 - \rho )^{2} }}{{(4 - 3\rho )^{2} }},\quad f_{4} = \frac{{16(1 - \rho )^{2} (2 - \rho )}}{{(4 - \rho )^{2} (4 - 3\rho )^{2} }},\quad f_{5} = \frac{{8(1 - \rho )^{2} (2 - \rho )}}{{(4 - \rho )(4 - 3\rho )^{2} }}. \\ \end{aligned}$$

Appendix 3

In this Appendix, we explain why the total demand for a given firm is constant in equilibrium. As the market is assumed to be covered, it is sufficient to show that the total number of consumers served by the two firms is equal in equilibrium.

When setting the price for a given product, each firm faces the usual value versus volume trade-off: Increasing the price of one product increases the profit from each unit of that product that the firm sells, but it also affects all products’ sales volume. In equilibrium, these two effects compensate each other. Hence, from the first-order conditions, the quantity that is sold of a given product is equal to the aggregate net loss in all products’ sales due to a marginal increase in the price of that product.

For the incumbent, this net loss in sales is due to the decrease in the sales of the product with a higher price plus the increase in (retail or wholesale) sales of other products. For the entrant, the net loss in sales results from the decrease in the product’s sales, but is compensated by a reduction in wholesale costs. The total output that is sold by each multi-product firm then corresponds to adding the net losses in the value of sales due to a unit increase in the prices of all its products.

If there were no input sales involved, firms would be symmetric, and each firm would have the same aggregate quantity. In fact, firm asymmetry in our model results solely from the fact that one firm sells an input to its competitor: The incumbent has wholesale revenues, whereas the entrant incurs wholesale costs. However, as is explained below, the effects of increasing the prices of their products on wholesale revenue or cost are the same.

When the incumbent marginally raises its prices, there is an increase in the wholesale revenue of the high-quality product that is given by \(w\left( {\frac{{\partial q_{EH} }}{{\partial p_{IH} }} + \frac{{\partial q_{EH} }}{{\partial p_{IL} }}} \right)\). As for the entrant, when it marginally raises retail prices, it faces a similar reduction in its input costs (instead of an increase in wholesale revenue). The increase in the high-quality product’s retail price reduces the input costs by \(\left( {w\frac{{\partial q_{EH} }}{{\partial p_{EH} }}} \right)\), and the increase in the low-quality product’s retail price increases the input costs by \(\left( {w\frac{{\partial q_{EH} }}{{\partial p_{EL} }}} \right)\) since some “firm-oriented” consumers switch from high-quality to low-quality products and vice versa.

Thus, the impact of an increase in the wholesale revenue due to an increase in the incumbent’s retail prices is given by \(w\left( {\frac{{\partial q_{EH} }}{{\partial p_{IH} }} + \frac{{\partial q_{EH} }}{{\partial p_{IL} }}} \right)\), which reflects the additional wholesale revenue from those consumers who switch from the incumbent to the entrant. This effect can be compared to the reduction in the entrant’s input costs due to an increase in its retail prices: \(w\left( {\frac{{\partial q_{EH} }}{{\partial p_{EH} }} - \frac{{\partial q_{EH} }}{{\partial p_{EL} }}} \right)\).

With \(\frac{{\partial q_{EH} }}{{\partial p_{EH} }} = \frac{{\partial q_{EH} }}{{\partial p_{IH} }} + \frac{{\partial q_{EH} }}{{\partial p_{EL} }} + \frac{{\partial q_{EH} }}{{\partial p_{IL} }}\), which in the general linear demand function \(D_{ij} = A_{j} - bp_{ij} + cp_{mj} + dp_{il} + ep_{ml}\) amounts to \(b = c + d + e\), the impact on the entrant’s equilibrium aggregate output of the wholesale activity is equal to the impact on the incumbent’s output. As the single source of asymmetry between the entrant and the incumbent impacts the two firms equally, they both sell the same quantity in equilibrium. This follows from: (1) demand symmetry; and (2) the fact that a similar increase in all prices leaves the quantity demanded unchanged (i.e., the market is fully covered).

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Brito, D., Tselekounis, M. On the Impact of Input Prices on an Entrant’s Profit Under Multi-Product Competition. Rev Ind Organ 50, 105–125 (2017). https://doi.org/10.1007/s11151-016-9531-2

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11151-016-9531-2

Keywords

JEL Classification

Navigation