TY - GEN KW - Supplier Konkurrenz KW - Ausschließendes PricingBilateral oligopoly KW - Buyer power KW - Single and dual sourcing KW - Supplier competition KW - Exclusionary pricing ID - heidok12566 Y1 - 2011/// TI - Strategic Sourcing to Sustain Supplier Competition - Theoretical and Experimental Analyses AV - public N2 - In a bilateral oligopoly where strategic buyers source a good from competing suppliers over time, buyers often have an incentive to sustain competition while suppliers benefit from excluding rival suppliers. We model this situation as a two period asymmetric cost Bertrand duopoly, where in the second period only those suppliers are active whose profit is sufficiently high in the first period. One or two buyers decide in each period how to source their demand from the suppliers. We solve the model analytically for a single buyer. In the first period the buyer dual sources and sustains competition when prices are high whereas for low prices single sourcing prevails. When the cost difference of suppliers is small, a unique subgame-perfect equilibrium in pure strategies exists. The more efficient supplier excludes the less efficient supplier by pricing at or below the cost of this supplier and the buyer sources all demand from him. With multiple identical buyers, equilibrium results do not change. Implementing the model numerically, we find that results are robust for coarse quantities. Exclusion can then occur even for large cost differences, while also equilibria exist where competition is often sustained. In addition, we implement an alternative procurement process where the buyer first sets the quantities to be sourced dependent on the relative prices. Then, equilibria where the buyer sources from both suppliers and competition is sustained with a high probability can exist even for small cost differences. The surplus of the buyer can, but need not, increase compared to the initial process. This model with its predicted predominance of exclusion is used in two experiments. First, we improve coordination from the baseline model with two independent buyers across treatments by allowing communication, introducing a single buyer and automating buyers. Outcomes in the second period are not observed to differ between treatments. In the first period, buyers nearly always dual source and sustain competition when this is rational for all treatments. With two independent buyers we observe exclusionary pricing to be predominant and accordingly competition is frequently not sustained. Contrary to the subgame-perfect prediction of no difference however, with improved coordination of buyers, pricing becomes more accommodating. Accordingly, competition is sustained more often and there is weak evidence that also buyer surplus increases in turn. In addition, we analyze experimentally the role of the procurement process with two buyers. Again, in the non-strategic setting of the second period, outcomes for all procurement processes are comparable. In the first period, supplier exclusion is predominant not only in the baseline process which corresponds to a procurement auction, but also for nonlinear prices in split-award auctions. Then, suppliers demand premia for small quantities which would allow to sustain competition, an option buyers however frequently do not use. When subsidies can be paid to suppliers or when buyers take their sourcing decisions before suppliers bid however, suppliers price more passively and competition is sustained substantially more often than in the baseline, in line with the prediction. Buyer sourcing is relatively close to rational behavior across procurement processes, although buyers are not always successful in sustaining competition when this is optimal. Therefore, the predicted increase in buyer surplus for subsidies and pre-announced sourcing is not observed. We thus find that both theoretically and experimentally, even with strong incentives for buyers to sustain competition, aggressive supplier behavior often entails exclusionary outcomes. However, both improved coordination of sourcing decisions and modifications of the procurement process are experimentally found to induce more accommodating pricing and lead to competition being sustained frequently. UR - https://archiv.ub.uni-heidelberg.de/volltextserver/12566/ A1 - Wilken, Florian ER -