Sourcing Co-Created Products: Should Your Suppliers Collaborate on Cost Reductions?


We study how a multi-product downstream firm should source from the upstream market—single-source versus multi-source—in a situation where the products are co-created with the suppliers. We conceptualize co-creation as investments that are made at the different hierarchical levels and that are aimed at reducing the production costs incurred by the supplies. In the case when the downstream firm sources its products from multiple suppliers, it can foster cost-reduction collaboration among the suppliers. We find that the downstream firm may be worse off when the upstream suppliers collaborate. For a commonly used additively separable cost function, we show that the downstream firm’s optimal strategy is multi-source co-creation without collaboration. Multi-sourcing softens the holdup problem and leads to a positive level of investment by the downstream firm.

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


  1. Burke, G. J., Carrillo, J. E., & Vakharia, A. J. (2007). Single versus multiple supplier sourcing strategies. European Journal of Operational Research, 182(1), 95–112.

    Article  Google Scholar 

  2. Calzolari, G., Felli, L., Koenen, J., Spagnolo, G., & Stahl, K. O. (2017). Relational contracts, competition and innovation: Theory and evidence from German car manufacturers. Working Paper.

  3. Coase, R. H. (2000). The acquisition of fisher body by general motors. Journal of Law and Economics, 43(1), 15–32.

    Article  Google Scholar 

  4. Cohen, L., & Young, A. (2006). Multisourcing: Moving beyond outsourcing to achieve growth and agility. Brighton: Harvard Business School Press.

    Google Scholar 

  5. Cohen, W. M., & Levinthal, D. A. (1990). Absorptive capacity: A new perspective on learning and innovation. Administrative Science Quarterly, 35(1), 128–152.

    Article  Google Scholar 

  6. Cohen-Vernik, D., Niladri, S., & Pazgal, A. (2018). Competing with co-created products. Working Paper.

  7. Feng, B. (2012). Multisourcing suppliers selection in service outsourcing. Journal of the Operational Research Society, 63(5), 582–596.

    Article  Google Scholar 

  8. Goyal, S., & Moraga-González, J. L. (2001). R&D networks. RAND Journal of Economics, 32(4), 686–707.

    Article  Google Scholar 

  9. Griffin, A., & Hauser, J. R. (1993). The voice of the customer. Marketing Science, 12(1), 1–27.

    Article  Google Scholar 

  10. Grout, P. A. (1984). Investment and wages in the absence of binding contracts: A Nash bargaining approach. Econometrica, 52(2), 449–460.

    Article  Google Scholar 

  11. Jaworski, B., & Kohli, A. K. (2006). Co-creating the voice of the customer. In R. F. Lusch & S. L. Vargo (Eds.), The service-dominant logic of marketing (pp. 109–117). New York: Routledge.

    Google Scholar 

  12. Kalaignanam, K., & Varadarajan, R. (2006). Customers as co-producers: Implications for marketing strategy effectiveness and marketing operations efficiency. In R. F. Lusch & S. L. Vargo (Eds.), The service-dominant logic of marketing (pp. 166–179). New York: Routledge.

    Google Scholar 

  13. Klein, B. (2000). Fisher-General Motors and the nature of the firm. Journal of Law and Economics, 43(1), 105–142.

    Article  Google Scholar 

  14. Kogut, B., & Zander, U. (1992). Knowledge of the firm, combinative capabilities and the replication of technologies. Organization Science, 3(3), 383–397.

    Article  Google Scholar 

  15. Levin, R. C., & Reiss, P. C. (1988). Cost-reducing and demand-creating R&D with spillovers. RAND Journal of Economics, 19(4), 538–556.

    Article  Google Scholar 

  16. Liker, J., & Choi, T. Y. (2004). Building deep supplier relationships. Harvard Business Review, 82, 104–113.

    Google Scholar 

  17. Moreau, C. P., Bonney, L., & Herd, K. B. (2011). It’s the thought (and the effort) that counts: How customizing for others differs from customizing for oneself. Journal of Marketing, 75(5), 120–133.

    Article  Google Scholar 

  18. Moreau, C. P., & Herd, K. B. (2010). To each his own? How comparisons with others influence consumers’ evaluations of their self-designed products. Journal of Consumer Research, 36(5), 806–819.

    Article  Google Scholar 

  19. Mukherjee, A., & Tsai, Y. (2013). Multi-sourcing as an entry deterrence strategy. International Review of Economics and Finance, 25, 108–112.

    Article  Google Scholar 

  20. Pack, H., & Saggi, K. (2001). Vertical technology transfer via international outsourcing. Journal of Development Economics, 65(2), 389–415.

    Article  Google Scholar 

  21. Payne, A. F., Storbacka, K., & Frow, P. (2008). Managing the co-creation of value. Journal of the Academy of Marketing Science, 36(1), 83–96.

    Article  Google Scholar 

  22. Randall, T., Terwiesch, C., & Ulrich, K. T. (2007). User design of customized products. Marketing Science, 26(2), 268–280.

    Article  Google Scholar 

  23. Redström, J. (2006). Towards user design? On the shift from object to user as the subject of design. Design Studies, 27(2), 123–139.

    Article  Google Scholar 

  24. Sawhney, M., Verona, G., & Prandelli, E. (2005). Collaborating to create: The Internet as a platform for customer engagement in product innovation. Journal of Interactive Marketing, 19(4), 4–17.

    Article  Google Scholar 

  25. Sen, D., & Tauman, Y. (2007). General licensing schemes for a cost-reducing innovation. Games and Economic Behavior, 59(1), 163–186.

    Article  Google Scholar 

  26. Spencer, B. J. (2005). International outsourcing and incomplete contracts. Canadian Journal of Economics, 38(4), 1107–1135.

    Article  Google Scholar 

  27. Shy, O., & Stenbacka, R. (2003). Strategic outsourcing. Journal of Economic Behavior and Organization, 50(2), 203–224.

    Article  Google Scholar 

  28. Syam, N., & Pazgal, A. (2013). Co-creation with production externalities. Marketing Science, 32(5), 805–820.

    Article  Google Scholar 

  29. Thomke, S., & von Hippel, E. (2002). Customers as innovators: A new way to create value. Harvard Business Review, 80(4), 74–81.

    Google Scholar 

  30. Tirole, J. (1986). Procurement and renegotiation. Journal of Political Economy, 94(2), 235–259.

    Article  Google Scholar 

  31. von Hippel, E. (1986). Lead-firms: A source of novel product concepts. Management Science, 32(7), 791–805.

    Article  Google Scholar 

  32. von Hippel, E., & Katz, R. (2002). Shifting innovation to firms via toolkits. Management Science, 48(7), 821–833.

    Article  Google Scholar 

  33. Wiener, M., & Saunders, C. (2014). Forced competition in IT multi‐sourcing. Journal of Strategic Information Systems, 23(3), 210–225.

    Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Oksana Loginova.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.



Proof of Lemma 1

Suppose that v is relatively high, so that in equilibrium all consumers buy (the market is fully covered). The ‘critical’ consumer is located at \(\hat{x}\) that satisfies

$$\begin{aligned} v-t\hat{x}-p_1= & {} v-t(1-\hat{x})-p_2,\\ \hat{x}= & {} \frac{t+p_2-p_1}{2t}. \end{aligned}$$

Given \(p_1^w\) and \(p_2^w\), F faces the following constrained optimization problem:

$$\begin{aligned} \max _{p_1, p_2} \hat{x}(p_1-p_1^w)+(1-\hat{x})(p_2-p_2^w)= & {} \frac{t+p_2-p_1}{2t}(p_1-p_1^w)\\&\quad +\frac{t+p_1-p_2}{2t}(p_2-p_2^w) \end{aligned}$$

subject to

$$\begin{aligned} v-t\hat{x}-p_1\ge 0. \end{aligned}$$

The constraint guarantees that the critical consumer buys the product; it holds with equality at the solution. Thus, we have

$$\begin{aligned} \max _{p_1, p_2} \frac{t+p_2-p_1}{2t}(p_1-p_1^w) +\frac{t+p_1-p_2}{2t}(p_2-p_2^w) \end{aligned}$$

subject to

$$\begin{aligned} v-\frac{t+p_1+p_2}{2}=0. \end{aligned}$$

Substituting the constraint into the objective function yields

$$\begin{aligned}\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\max _{p_1} \frac{v-p_1}{t}(p_1-p_1^w)+\frac{t-v+p_1}{t}(2v-t-p_1-p_2^w),\end{aligned}$$
$$\begin{aligned}\text{ F.O.C. } \quad \frac{1}{t}\left( -p_1+p_1^w+v-p_1+2v-t-p_1-p_2^w-t+v-p_1\right) =0,\end{aligned}$$
$$\begin{aligned}\quad\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\, -4p_1+4v-2t+p_1^w-p_2^w=0, \end{aligned}$$


$$\begin{aligned} p_1=v-\frac{t}{2}+\frac{p_1^w-p_2^w}{4}, \end{aligned}$$


$$\begin{aligned} p_2=v-\frac{t}{2}+\frac{p_2^w-p_1^w}{4} \end{aligned}$$


$$\begin{aligned} \hat{x}= \frac{1}{2}-\frac{p_1^w-p_2^w}{4t}. \end{aligned}$$

Proof of Lemma 2

S1 solves

$$\begin{aligned}&\max _{p_1^w} \hat{x}(p_1^w-c_1)= \left( \frac{1}{2} -\frac{p_1^w-p_2^w}{4t}\right) (p_1^w-c_1),\\&\,\,\,\,\,\,\,\text{ F.O.C. } \quad \frac{1}{2}-\frac{p_1^w-p_2^w}{4t} -\frac{p_1^w-c_1}{4t}=0. \end{aligned}$$

Similarly, S2 solves

$$\begin{aligned}&\max _{p_2^w} (1-\hat{x})(p_2^w-c_2)=\left( \frac{1}{2} -\frac{p_2^w-p_1^w}{4t}\right) (p_2^w-c_2),\\&\,\,\,\,\,\,\,\,\,\,\,\,\,\,\text{ F.O.C. } \quad \frac{1}{2}-\frac{p_2^w-p_1^w}{4t} -\frac{p_2^w-c_2}{4t}=0. \end{aligned}$$

We put the two first-order conditions into a system of equations and solve for \(p_1^w\) and \(p_2^w\):

$$\begin{aligned} p_1^w=\frac{6t+2c_1+c_2}{4}\quad \text{ and } \quad p_2^w=\frac{6t+2c_2+c_1}{4}. \end{aligned}$$

Proof of Lemma 3

We use the results of Lemmas 1 and 2 to obtain the continuation profits of S1, S2, and F as functions of \(c_1\) and \(c_2\):

$$\begin{aligned} \pi ^{ms}_1(c_1, c_2)&=\left( \frac{1}{2}-\frac{p_1^w-p_2^w}{4t}\right) (p_1^w-c_1)\\&=\left( \frac{1}{2}-\frac{c_1-c_2}{16t}\right) \frac{6t-2c_1+c_2}{4} =\frac{1}{64t}(8t-c_1+c_2)(6t-2c_1+c_2). \end{aligned}$$


$$\begin{aligned} \pi ^{ms}_2(c_1, c_2)=\frac{1}{64t}(8t-c_2+c_1)(6t-2c_2+c_1). \end{aligned}$$


$$\begin{aligned} \pi ^{ms}_F(c_1,c_2)&= \hat{x}(p_1-p_1^w)+(1-\hat{x})(p_2-p_2^w)\\&= {} \left( \frac{1}{2}-\frac{p_1^w-p_2^w}{4t}\right) \left( v-\frac{t}{2} +\frac{p_1^w-p_2^w}{4}-p_1^w\right) \\&+\left( \frac{1}{2}+\frac{p_1^w-p_2^w}{4t}\right) \left( v-\frac{t}{2} -\frac{p_1^w-p_2^w}{4}-p_2^w\right) \\&= {} v-\frac{t}{2}-\frac{p_1^w+p_2^w}{2}+\frac{(p_1^w-p_2^w)^2}{8t}\\&= v-2t-\frac{3(c_1+c_2)}{8}+\frac{(c_1-c_2)^2}{128t}. \end{aligned}$$

Proof of Lemma 4

S1 solves

$$\begin{aligned} \max _{I_1} \frac{1}{64t}(8t-c(e,I_1+\gamma I_2)+c(e,I_2+\gamma I_1))(6t-2c(e,I_1+\gamma I_2) +\,c(e,I_2+\gamma I_1))-I_1, \end{aligned}$$
$$\begin{aligned}&\text{ F.O.C. } \quad \frac{1}{64t}\left( \left( -\frac{\partial c(e,I_1+\gamma I_2)}{\partial I}+\gamma \frac{\partial c(e,I_2+\gamma I_1)}{\partial I}\right) (6t-2c(e,I_1+\gamma I_2)\right. +\,c(e,I_2+\gamma I_1))\\&\quad +\,(8t-c(e,I_1+\gamma I_2)\left. +\,c(e,I_2+\gamma I_1)) \left( -2\frac{\partial c(e,I_1+\gamma I_2)}{\partial I} +\gamma \frac{\partial c(e,I_2+\gamma I_1)}{\partial I}\right) \right) =1. \end{aligned}$$

S2 solves

$$\begin{aligned}\max _{I_2} \frac{1}{64t}(8t-c(e,I_2+\gamma I_1) +c(e,I_1+\gamma I_2))(6t-2c(e,I_2+\gamma I_1) +\,c(e,I_1+\gamma I_2))-I_2, \end{aligned}$$
$$\begin{aligned}&\text{ F.O.C. } \quad \frac{1}{64t} \left( \left( -\frac{\partial c(e,I_2+\gamma I_1)}{\partial I}+\gamma \frac{\partial c(e,I_1+\gamma I_2)}{\partial I}\right) (6t-2c(e,I_2+\gamma I_1)\right. \\&\quad +\,c(e,I_1+\gamma I_2)) +\,(8t-c(e,I_2+\gamma I_1)\left. +\,c(e,I_1+\gamma I_2)) \left( -2\frac{\partial c(e,I_2+\gamma I_1)}{\partial I}+\gamma \frac{\partial c(e,I_1+\gamma I_2)}{\partial I}\right) \right) =1. \end{aligned}$$

F solves

$$\begin{aligned}\max _{e} v-2t-\frac{3(c(e, I_1+\gamma I_2)+c(e,I_2+\gamma I_1))}{8} +\frac{(c(e, I_1+\gamma I_2)-c(e, I_2+\gamma I_1))^2}{128t}-e, \end{aligned}$$
$$\begin{aligned} &\text{ F.O.C. } \quad -\frac{3}{8} \left( \frac{\partial c(e, I_1+\gamma I_2)}{\partial e} +\frac{\partial c(e, I_2+\gamma I_1)}{\partial e}\right) \\&\quad +\,\frac{(c(e, I_1+\gamma I_2)-c(e, I_2+\gamma I_1))}{64t} \left( \frac{\partial c(e, I_1+\gamma I_2)}{\partial e} -\frac{\partial c(e, I_2+\gamma I_1)}{\partial e}\right) =1. \end{aligned}$$

We put the three first-order conditions into a system of equations to solve for the equilibrium values of \(I_1\), \(I_2\), and e. In equilibrium S1 and S2 will choose the same level of investment; we denote it by \(I^C\). Let \(e^C\) denote the equilibrium level of F’s investment. \(I^C\) and \(e^C\) must satisfy

$$\left\{ \begin{array}{l} -\dfrac{3}{4}\dfrac{\partial c(e^C, I^C+\gamma I^C)}{\partial e} = 1 \\ -\dfrac{1}{64t}(22t-14\gamma t-(1-\gamma )c(e^C, I^C+\gamma I^C)) \dfrac{\partial c(e^C, I^C+\gamma I^C)}{\partial I} = 1. \end{array}\right.$$

Proof of Proposition 1

It follows from Lemma 3 that \(\tilde{I}^C\) and \(e^C\) are implicitly defined by

$$\left\{ \begin{array}{l} -\dfrac{3}{4}\dfrac{\partial c(e^C, \tilde{I}^C)}{\partial e} = 1 \\ -\dfrac{1}{64t}(22t-14\gamma t-(1-\gamma )c(e^C, \tilde{I}^C)) \dfrac{\partial c(e^C,\tilde{I}^C)}{\partial I} = 1. \end{array}\right.$$

Taking the total derivative with respect to \(\gamma \) yields

$$\left\{ \begin{array}{l} -\dfrac{3}{4}\left( \dfrac{\partial ^2 c}{\partial e^2} \dfrac{\mathrm{d} e^C}{\mathrm{d} \gamma } +\dfrac{\partial ^2 c}{\partial e \partial I} \dfrac{\mathrm{d} \tilde{I}^C}{\mathrm{d} \gamma }\right) = 0 \\ -\dfrac{1}{64t}\left( \left( -14t+c+\gamma \left( \dfrac{\partial c}{\partial e}\dfrac{\mathrm{d} e^C}{\mathrm{d} \gamma } +\dfrac{\partial c}{\partial I}\dfrac{\mathrm{d} \tilde{I}^C}{\mathrm{d} \gamma }\right) \right) \dfrac{\partial c}{\partial I}\right. \\ \quad \quad \left. +(22t-14\gamma t-(1-\gamma )c) \left( \dfrac{\partial ^2 c}{\partial e \partial I} \dfrac{\mathrm{d} e^C}{\mathrm{d} \gamma }+\dfrac{\partial ^2 c}{\partial I^2} \dfrac{\mathrm{d} \tilde{I}^C}{\mathrm{d} \gamma }\right) \right) =0. \end{array}\right.$$

This is a linear system of equations with respect to \(\mathrm{d} e^C/\mathrm{d} \gamma \) and \(\mathrm{d}\tilde{I}^C/\mathrm{d} \gamma \):

$$\left\{ \begin{array}{l} \dfrac{\partial ^2 c}{\partial e^2}\dfrac{\mathrm{d} e^C}{\mathrm{d} \gamma } +\dfrac{\partial ^2 c}{\partial e \partial I}\dfrac{\mathrm{d} \tilde{I}^C}{\mathrm{d} \gamma }= 0 \\ \left( \gamma \dfrac{\partial c}{\partial e}\dfrac{\partial c}{\partial I} +(22t-14\gamma t-(1-\gamma )c)\dfrac{\partial ^2 c}{\partial e \partial I}\right) \dfrac{\mathrm{d} e^C}{\mathrm{d} \gamma }\\ \quad \quad +\left( \gamma \left( \dfrac{\partial c}{\partial I}\right) ^2 +(22t-14\gamma t-(1-\gamma )c)\dfrac{\partial ^2 c}{\partial I^2}\right) \dfrac{\mathrm{d} \tilde{I}^C}{\mathrm{d} \gamma } =(14t-c)\dfrac{\partial c}{\partial I}. \end{array}\right. $$

Recall that \(\partial c/\partial e<0\), \(\partial c/\partial I<0\), \(\partial ^2 c/\partial e^2>0\), \(\partial ^2 c/\partial I^2>0\), and \(\partial ^2 c/\partial e \partial I>0\). We can rewrite our system of equations as

$$\begin{aligned} \left\{ \begin{array}{l} C_1\dfrac{\mathrm{d} e^C}{\mathrm{d} \gamma }+C_2\dfrac{\mathrm{d} \tilde{I}^C}{\mathrm{d} \gamma }= 0 \\ C_3\dfrac{\mathrm{d} e^C}{\mathrm{d} \gamma }+C_4\dfrac{\mathrm{d} \tilde{I}^C}{\mathrm{d} \gamma } =-C_5, \end{array}\right. \end{aligned}$$


$$\begin{aligned} C_1= & {} \frac{\partial ^2 c}{\partial e^2},\\ C_2= & {} \frac{\partial ^2 c}{\partial e \partial I}, \end{aligned}$$
$$\begin{aligned}C_3= & {} \gamma \frac{\partial c}{\partial e}\frac{\partial c}{\partial I} +(22t-14\gamma t-(1-\gamma )c)\frac{\partial ^2 c}{\partial e \partial I},\\ C_4= & {} \gamma \left( \frac{\partial c}{\partial I}\right) ^2 +(22t-14\gamma t-(1-\gamma )c)\frac{\partial ^2 c}{\partial I^2}, \end{aligned}$$


$$\begin{aligned} C_5=-(14t-c)\frac{\partial c}{\partial I} \end{aligned}$$

are all positive. Solving the system of equations for \(\mathrm{d} e^C/\mathrm{d} \gamma \) and \(\mathrm{d} \tilde{I}^C/\mathrm{d} \gamma \) yields

$$\begin{aligned} \dfrac{\mathrm{d} e^C}{\mathrm{d} \gamma }=\frac{C_2 C_5}{C_1 C_4-C_2 C_3} \end{aligned}$$


$$\begin{aligned} \dfrac{\mathrm{d} \tilde{I}^C}{\mathrm{d} \gamma }=-\frac{C_1 C_5}{C_1 C_4-C_2 C_3}. \end{aligned}$$

Therefore, when \(C_1 C_4-C_2 C_3>0\), \(\mathrm{d} e^C/\mathrm{d} \gamma \) is positive, and \(\mathrm{d} \tilde{I}^C/\mathrm{d} \gamma \) is negative. Otherwise, \(\mathrm{d} e^C/\mathrm{d} \gamma \) is negative, and \(\mathrm{d} \tilde{I}^C/\mathrm{d} \gamma \) is positive.

Proof of Lemma 6

When \(p_1^w\) and \(p_2^w\) are high, F solves

$$\begin{aligned} \max _{p_1, p_2} \hat{x}_1(p_1-p_1^w)+ (1-\hat{x}_2)(p_2-p_2^w), \end{aligned}$$


$$\begin{aligned} v-p_1-t\hat{x}_1=0 \quad \Rightarrow \quad \hat{x}_1=\frac{v-p_1}{t} \end{aligned}$$


$$\begin{aligned} v-p_2-t(1-\hat{x}_2)=0 \quad \Rightarrow \quad 1-\hat{x}_2=\frac{v-p_2}{t}. \end{aligned}$$

Thus, we have

$$\begin{aligned} \max _{p_1, p_2} \frac{v-p_1}{t}(p_1-p_1^w)+ \frac{v-p_2}{t}(p_2-p_2^w). \end{aligned}$$

Differentiating with respect to \(p_1\) and setting the derivative to zero yield

$$\begin{aligned} p_1=\frac{v+p_1^w}{2}. \end{aligned}$$

Differentiating with respect to \(p_2\) and setting the derivative to zero yield

$$\begin{aligned} p_2=\frac{v+p_2^w}{2}. \end{aligned}$$


$$\begin{aligned} \hat{x}_1=\frac{v-p_1^w}{2t} \quad \text{ and } \quad 1-\hat{x}_2=\frac{v-p_2^w}{2t}. \end{aligned}$$

In equilibrium, S sets wholesale prices \(p_1^w\) and \(p_2^w\) just low enough so that all consumers purchase in equilibrium. (Recall that v has been assumed to be sufficiently high, implying high margins.) That is, \(\hat{x}_1\) and \(1-\hat{x}_2\) must sum up to one:

$$\begin{aligned} \frac{v-p_1^w}{2t}+\frac{v-p_2^w}{2t}=1, \end{aligned}$$


$$\begin{aligned} p_1^w+p_2^w=2v-2t \quad \Rightarrow \quad p_2^w=2v-2t-p_1^w. \end{aligned}$$

Next, given \(c_1\) and \(c_2\), S solves

$$\begin{aligned} \max _{p_1^w, p_2^w} \frac{v-p_1^w}{2t}(p_1^w-c_1)+\frac{v-p_2^w}{2t}(p_2^w-c_2) \end{aligned}$$

subject to

$$\begin{aligned} p_2^w=2v-2t-p_1^w. \end{aligned}$$

Substituting the constraint into the objective function yields

$$\begin{aligned}\,\,\,\,\,\,\,\,\,\,\,\,\max _{p_1^w} \frac{v-p_1^w}{2t}(p_1^w-c_1)+\frac{2t-v+p_1^w}{2t}(2v-2t-p_1^w-c_2), \end{aligned}$$
$$\begin{aligned}\mathrm{F.O.C.} \quad \frac{1}{2t}(-p_1^w+c_1+v-p_1^w+2v-2t-p_1^w-c_2-2t+v-p_1^w)=0, \end{aligned}$$
$$\begin{aligned}\quad \,\,\,\,\,\,\,\,\,\,\,\,-4p_1^w+4v-4t+c_1-c_2=0, \end{aligned}$$
$$\begin{aligned}\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,p_1^w=v-t+\frac{c_1-c_2}{4}, \end{aligned}$$


$$\begin{aligned} p_2^w=v-t+\frac{c_2-c_1}{4}. \end{aligned}$$

Finally, we derive S’s continuation profit as a function of \(c_1\) and \(c_2\):

$$\begin{aligned} \pi ^{ss}_S(c_1,c_2)= & {} \frac{v-p_1^w}{2t}(p_1^w-c_1)+\frac{v-p_2^w}{2t}(p_2^w-c_2)\\= & {} \left( \frac{1}{2}-\frac{c_1-c_2}{8t}\right) \left( v-t+\frac{c_1-c_2}{4}-c_1\right) +\left( \frac{1}{2}+\frac{c_1-c_2}{8t}\right) \left( v-t-\frac{c_1-c_2}{4}-c_2\right) \\= & {} v-t-\frac{c_1+c_2}{2}+\frac{(c_1-c_2)^2}{16t}. \end{aligned}$$

F’s continuation profit as a function of \(c_1\) and \(c_2\) is:

$$\begin{aligned} \pi ^{ss}_F(c_1,c_2)&=\frac{v-p_1}{t}(p_1-p_1^w)+ \frac{v-p_2}{t}(p_2-p_2^w) =\frac{(v-p_1^w)^2}{4t}+\frac{(v-p_2^w)^2}{4t}\\&=\frac{1}{4t}\left( \left( t-\frac{c_1-c_2}{4}\right) ^2 +\left( t+\frac{c_1-c_2}{4}\right) ^2\right) \\&=\frac{1}{4t}\left( 2t^2+2\left( \frac{c_1-c_2}{4}\right) ^2\right) =\frac{t}{2}+\frac{(c_1-c_2)^2}{32t}. \end{aligned}$$

Proof of Proposition 3

F solves

$$\begin{aligned} \max _{e} \frac{t}{2}+\frac{(c(e,I_1+\delta I_2) -c(e,I_2+\delta I_1))^2}{32t}-e. \end{aligned}$$

The higher is the difference in the production costs, the higher is F’s profit. Suppose for definitiveness \(I_2>I_1\). Then

$$\begin{aligned} |c(e,I_1+\delta I_2)-c(e,I_2+\delta I_1)| =c(e,I_1+\delta I_2)-c(e,I_2+\delta I_1) \end{aligned}$$


$$\begin{aligned} \frac{\partial }{\partial e}(c(e,I_1+\delta I_2)-c(e,I_2+\delta I_1)) =-\int _{I_1}^{I_2} \frac{\partial ^2 c(e,I)}{\partial e \partial I} \mathrm{d}I<0. \end{aligned}$$

It follows that zero investment is F’s dominant strategy.

Proof of Lemma 7

S solves

$$\begin{aligned}&\max _{I_1, I_2} \frac{v}{2}-\frac{t}{2}-\frac{c(0,I_1+\delta I_2) +c(0,I_2+\delta I_1)}{2} +\frac{(c(0,I_1+\delta I_2) -c(0,I_2+\delta I_1))^2}{16t}-I_1-I_2,\\&\,\,\,\,\,\,\,\left\{ \begin{array}{l} \dfrac{1}{2}\left( \dfrac{\partial c(0,I_1+\delta I_2)}{\partial I} +\delta \dfrac{\partial c(0,I_2+\delta I_1)}{\partial I}\right) \\ \quad \quad +\dfrac{(c(0,I_1+\delta I_2)-c(0,I_2+\delta I_1))}{8t} \left( \dfrac{\partial c(0,I_1+\delta I_2)}{\partial I}-\delta \dfrac{\partial c(0,I_2+\delta I_1)}{\partial I}\right) =1\\ \dfrac{1}{2}\left( \dfrac{\delta \partial c(0,I_1+\delta I_2)}{\partial I} +\dfrac{\partial c(0,I_2+\delta I_1)}{\partial I}\right) \\ \quad \quad +\dfrac{(c(0,I_1+\delta I_2)-c(0,I_2+\delta I_1))}{8t} \left( \delta \dfrac{\partial c(0,I_1+\delta I_2)}{\partial I} -\dfrac{\partial c(0,I_2+\delta I_1)}{\partial I}\right) =1. \end{array}\right. \end{aligned}$$

Depending on the parameters t and \(\delta \), as well as the functional from of the cost function, we can obtain a symmetric solution, or two asymmetric ones. We assume that \(\delta \) is sufficiently close to one, so that the solution is symmetric: \(I_1=I_2=I^*\). \(I^*\) must satisfy

$$\begin{aligned} -\frac{1}{2}(1+\delta )\frac{\partial c(0,I^*+\delta I^*)}{\partial I}=1. \end{aligned}$$

Proof of Lemma 8

S1 solves

$$\begin{aligned}&\max _{I_1} \frac{1}{64t}(8t-c(0, I_1)+c(0, I_2))(6t-2c(0, I_1)+c(0, I_2))-I_1,\\&\,\,\,\,\text{ F.O.C. } \quad \frac{1}{64t}\left( -\frac{\partial c(0,I_1)}{\partial I}(6t-2c(0,I_1)+c(0,I_2))\right. \\&\,\,\,\,\,\,\,\,\,\,\,\quad + \left. (8t-c(0,I_1)+c(0,I_2)) \left( -2\frac{\partial c(0,I_1)}{\partial I}\right) \right) =1. \end{aligned}$$

S2 solves

$$\begin{aligned}&\max _{I_2} \frac{1}{64t}(8t-c(0,I_2)+c(0,I_1))(6t-2c(0,I_2)+c(0,I_1))-I_2,\\&\,\,\,\,\text{ F.O.C. } \quad \frac{1}{64t}\left( -\frac{\partial c(0,I_2)}{\partial I}(6t-2c(0,I_2)+c(0,I_1))\right. \\&\,\,\,\,\,\,\,\,\,\,\,\,\,\quad + \left. (8t-c(0,I_2)+c(0,I_1)) \left( -2\frac{\partial c(0,I_2)}{\partial I}\right) \right) =1. \end{aligned}$$

In equilibrium S1 and S2 will choose the same level of investment; we denote it by \(I^{\dagger }\). \(I^{\dagger }\) must satisfy

$$\begin{aligned} -\frac{1}{64t}(22t-c(0, I^{\dagger }))\frac{\partial c(0, I^{\dagger })}{\partial I} = 1. \end{aligned}$$

Proof of Proposition 4

With \(c(e,I)=\bar{c}-a\sqrt{e}-b\sqrt{I}\), the system of equations in Lemma 5 becomes

$$\begin{aligned}&\left\{ \begin{array}{l} \dfrac{3}{4}\times \dfrac{a}{2\sqrt{e^{NC}}} = 1 \\ \dfrac{1}{64t}\left( 22t-\left( \bar{c}-a\sqrt{e^{NC}}-b\sqrt{I^{NC}}\right) \right) \dfrac{b}{2\sqrt{I^{NC}}}= 1, \end{array}\right. \\ &\,\,\,\,\,\left \{ \begin{array}{l} \sqrt{e^{NC}}=\dfrac{3a}{8}\\ \left. 22t-\bar{c}+a\sqrt{e^{NC}}+b\sqrt{I^{NC}}\right) =\dfrac{128t}{b}\sqrt{I^{NC}}, \end{array}\right.\end{aligned}$$


$$\begin{aligned} e^{NC}=\left( \frac{3a}{8}\right) ^2 \end{aligned}$$


$$\begin{aligned} I^{NC}=\left( \frac{22t-\bar{c}+a\sqrt{e^{NC}}}{128t/b-b}\right) ^2. \end{aligned}$$

Setting \(e^{NC}\) to zero, we obtain

$$\begin{aligned} I^{\dagger }=\left( \frac{22t-\bar{c}}{128t/b-b}\right) ^2. \end{aligned}$$

Observe that \(I^{NC}>I^{\dagger }\). It immediately follows that strategy A without collaboration dominates strategy C. Indeed, the downstream firm F can choose \(e=0\) under strategy A without collaboration and get a higher profit than under strategy C. Since \(e=e^{NC}\) is F’s best response to \(I_1=I_2=I^{NC}\), F’s equilibrium profit under strategy A without collaboration is even higher.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Loginova, O., Syam, N.B. Sourcing Co-Created Products: Should Your Suppliers Collaborate on Cost Reductions?. Rev Ind Organ 56, 329–355 (2020).

Download citation


  • Co-creation
  • Collaboration
  • Holdup
  • Product sourcing

JEL Classification

  • C72
  • D4
  • L1
  • M31