TY - JOUR

T1 - Long-term Relationships

T2 - Static Gains and Dynamic Inefficiencies

AU - Hémous, David

AU - Olsen, Morten

PY - 2017/7/27

Y1 - 2017/7/27

N2 - In the 1980s the Japanese “keiretsu” system of interconnected business groups was praised as a model to emulate, but since then Japan has often been criticized for being less innovative than the United States. In this paper we connect the two views and argue that tight business relationships can create dynamic inefficiencies and reduce broad innovations. In particular, we consider the repeated interaction between final good producers and intermediate input suppliers, where the provision of the intermediate input is noncontractible. We build a cooperative equilibrium where producers can switch suppliers and start cooperation immediately with new suppliers. We first consider broad innovations: every period, one supplier has the opportunity to create a higher quality input that can be used by all producers. Since relationships are harder to break in the cooperative equilibrium the market size for potential innovators is smaller and the rate of innovation might be lower than in the noncooperative equilibrium. We contrast this with a setting with relationship-specific innovations that we show are encouraged by the establishment of relational contracts. We illustrate the predictions of the model using the recent business history of the United States and Japan and further use patent data to show that U.S. patents are more general than Japanese and even more so in sectors using more differentiated inputs.

AB - In the 1980s the Japanese “keiretsu” system of interconnected business groups was praised as a model to emulate, but since then Japan has often been criticized for being less innovative than the United States. In this paper we connect the two views and argue that tight business relationships can create dynamic inefficiencies and reduce broad innovations. In particular, we consider the repeated interaction between final good producers and intermediate input suppliers, where the provision of the intermediate input is noncontractible. We build a cooperative equilibrium where producers can switch suppliers and start cooperation immediately with new suppliers. We first consider broad innovations: every period, one supplier has the opportunity to create a higher quality input that can be used by all producers. Since relationships are harder to break in the cooperative equilibrium the market size for potential innovators is smaller and the rate of innovation might be lower than in the noncooperative equilibrium. We contrast this with a setting with relationship-specific innovations that we show are encouraged by the establishment of relational contracts. We illustrate the predictions of the model using the recent business history of the United States and Japan and further use patent data to show that U.S. patents are more general than Japanese and even more so in sectors using more differentiated inputs.

KW - Faculty of Social Sciences

KW - C73

KW - K12

KW - L14

KW - O31

KW - O43

U2 - 10.1093/jeea/jvx019

DO - 10.1093/jeea/jvx019

M3 - Journal article

VL - 16

SP - 383

EP - 435

JO - Journal of the European Economic Association

JF - Journal of the European Economic Association

SN - 1542-4774

IS - 3

ER -